Radar and guidance questions

Been reading my newly purchased copy of "British Secret Projects; Hypersonics, Ramjets & Missiles" and some questions occure to me.

S/Ldr Poole's analysis was I think being rather negative since you can't learn anything if you don't try and we where certainly trying. Though I'm not sure we got much chance to use the knowledge gained, leastways on AAMs.

Presumably I shall be blasted for daring to ask such questions.

Seems the UK had problems integrating guidance illuminators with its radars in the 1950s.
Two frequencies of illuminators where explored, X-band, and Q-band, an third was proposed in J-band but I'll come back to that one in a bit.

X-band seems the favoured, on the Red Dean AAM and on the Red Hebe AAM in virtualy all the designes even into the 1960s with a all-new AAM for the OR346 fighter (which was quite unusual in design), though GEC did develope a Q-band illuminator not sure if it worked well.

Was Vickers right? I seem to reccal Sparrow used an X-band illumnator on the AWG.10 used in F4's.

Seems they could'nt integrate it with the AI.23 monopulse set at all, it being unsuitable for CW illuminators, and had to concentrate on the AI.18 set instead. However, there they also experienced a lot of problems, signal polarisation and interfacing the CW illuminator with the pulsed radar.
Is this a product of the actual radar or just their inexperience with what was for them a new technology?

Clearly the radars where less than sparkling performers, but it does seem odd the great white hope of the AI.18 never materialised. Vickers worked rather hard to make it work, but what I can't see is whether they had any hope or whether it was a pointless effort and a new radar was required to get somewhere.

I know in the 60's a new AI.25 set was proposed for the Sea Vixen to extend its useful life. Possibly this was also related to future fighters, but I'm not sure its the same set as the Aspinal CW set the RN wanted.

Fairey prior to the '57 axe falling however proposed a SARH guided AAM using a J-band illuminator, which seems a very good basis for such a weapon, using as it did a continuous rod warhead, it seems to ahve been based on knowledge gained from the poorly performing Fireflash AAM.
Was this use of J-band any better a solution for illumination?

The AI-23 was a good radar design, and was developed for many years. As I understand it, the problems with introducing a CW illuminator into it were:

1) You need a CW transmitter approximately the same size as the pulse one used already. This means extra space, weight, and power usage. On the Lightning the whole radar was very tightly packed into a bullet shaped housing, and there was no room for any additional boxes.

2) You need either to multiplex the CW signal into the existing antenna system, or provide a separate antenna for the CW signal. If the CW signal is a different band from the main radar signal, a single antenna isn't going to be possible. The separate antenna approach was considered in the AI-23 case for example.

Either way, its not at all impossible, but its obviously easier to integrate from the beginning. If the AI-23 had been fitted to a larger airframe, for example the P.1121, it would have been more possible to integrate CW illumination.

In the case of AI-18, on Sea Vixen, space concerns were less of a problem, but this radar was older technology than AI-23 and has a worse reputation.

That would make sense, as the CAP Buccaneer design seems to be toting a larger AI.23 set from what I've read. From measurements on the BSP.4 picture I suspect it might even be a 36" dish though more likely a 33".
That of course was to potentialy be armed with Radar Red Top or Blue Dolphin (previously?).

Buccaneer would certainly have the potential space for the extra equipment, even if it has to move other equipment elsewhere in the airframe.
Good question, what is the radar in the B109 offering to Canada?

Same for the P1121. Though the BSP.Fighters book's image of the orriginal shows a rather small scanner, which must impact the design. The later 1960 offering to OR.339 shown in BSP.Bombers looks more appropriate.

It seems to me reading BSP:HRM that having got themselves a serviceable heat-homing AAM (Firestreak/Red Top), the Powers That Be were content to stick with it and not seriously develop radar guidance technology at all once Red Dean/Red Hebe had fallen by the wayside. Yet simultaneously we have the SAM constructors producing passable SARH-guided missiles (esp. Bloodhound) and the radars to work with them, and NONE of that knowledge is being shared with the struggling AAM manufacturers.

I know the problems are not identical - a SAM designed for a fifty-mile kill against a bomber at 40,000-plus feet doesn't need to worry so much about surface returns, for example, and you probably can't simply shrink Bloodhound's seeker and shove it in Red Top's nose, or Bloodhound's tracking/illuminating combination and shove it in Sea Vixen's/Javelin's (let alone Lightning's) - but surely things could have been made easier for the RAF and the aircraft manufacturers.

What was really needed, of course, was an ADV TSR.2, which might have provided a better launch platform for SARH AAMs (and an improved carriage platform for the radar system), and which (being for Britain's defence) might even have been a much harder beast to kill politically. But that would in turn have needed a search/track radar and illuminator (in the place of the surface attack system and "TFR illuminator") that could be slotted in with relatively little redesign effort and a missile to go with it, and that would in turn have depended on the long-term survival of an at least half-decent research effort into SARH missiles (TSR2 with a Red Hebe/Red Top mix?). Round and round and round and...

Ok, sorry to dig this thread up from the dead, but I was wondering if there are any details anywhere regarding the AI-25 or the Aspinal radars? I can not seem to find any information on British radar technology after the AI-23. I am assuming that there were at least proposals for radars for the Vickers Type-581/583 and the P.1154 when they were being considered?

Thanks in advance sealordlawrence.

sealordlawrence said:

I can not seem to find any information on British radar technology after the AI-23.

Click to expand...

AI-24 for the Tornado ADV is the public face, of course. But what went in between?

The failure of the SAM radar manufacturers to pool their knowledge with the AAM people always mystified me as well. Of course there must have been SOME AAM research going on all that time, because suddenly GEC - which failed to measure up w.r.t. Red Dean and Red Hebe - turns out essentially the best SARH AAM in the world (YMMV if you work for Selenia!).

There's not been enough written on the in-between projects - things like Fairey's SARH missile, Vickers' "Small Weapon" and the OR.329-matched snap-up missile (Forbat's theoretical study) and so on - nor on what went before. I'd give my eye teeth, for example, for a detailed story of Folland's struggles with Red Dean. IIRC, BSP4 mentions E.K. Coles working on the guidance system for the Folland version of the missile, but their otherwise very detailed website - though informative about Fireflash - gives not a squeak on the matter unless Buttler and Gibson are talking about unexplored or undescribed functions for AI.20 (later Red Steer).

I'm going to bump this up and reply to that last post which I hadd'nt noticed before.

Gut instinct tells me that turning the AI.20 into a radar to guide a SARH AAM is plausable. Precisely because the set already has a tight beam function in it for the beam rider missile, Fireflash.
In theory a larger dish and improvements to the set (power for one thing) where achievable during the correct timeframe. Keeping the beam on target is an interesting issue though since the trials sets where fixed to the planes axis for beam guidance.

Certainly I'd love to hear more about the Aspinal CW set (there must be some ministery documents if nothing else).

And similar for the E.K. Coles effort earlier.

However in the 60's we have at least two offerings of a SARH seeker for Red Top, so presumably they felt that giving the AI.23 an illuminator was by then achievable/affordable.
That might all make sense if seen in the context of the then aircraft proposals such AW.406.

I am going to add some thoughts and the little bit of information I have been able to glean from the sources currently at my disposal. To summarise, British A2A radar seems to have been as much of an abortion in this timeframe as the A2A guided weapons industry.

Firstly, the lineage:

AI.17: First designed in 1944 this did not make into production until 1955
AI.18: Appears to have been under development by 1951 but was not production ready until at least 1959
AI.19: ???
AI.20: Limited capability set as a backup for AI.23, used with Sky Flash and eventually morphed into Red Steer
AI.21: Westinghouse AN/APS-57 radar supplied under the US Military Assistance Program
AI.22: AN/APG-43
AI.23: As used on the lightning, further developed for Buccaneer as Blue Parrot and then even further into the TFR for TSR-2
AI.24: Fox Hunter
AI.25: The curious Sea Vixen upgrade suggestion, seems to have only ever existed as a scribbled note, though there are various references to 'lightened' AI.18 variants details are scant at best.

There are some curious diversions, Hawker were proposing J-Band AI.23 with (possibly) an additional Q-Band for target illumination (in a jamming environment?) to work with the Fairey Semi-Active radar weapon (which was itself J band). I assume that the reason for suggesting the Fairey weapon was that the original P.1121 proposal fell between the cancellation of the Blue Jay Mk5 and its resurrection as Blue Dolphin. Had P.1121 (or the EE P.8 for that matter) come to be I suspect that the Red Top / Blue Dolphin combination would have been used along with X band AI.23. Although interestingly, Overscan states that from reading the National Archives Red Hebe, in its dying days, was proposed with a J-Band monopulse seeker to better equip it against jamming. Furthermore it is stated in BSP4 that the radar for the F.155T aircraft would have almost certainly been the J-band AI-23 so J-band was really in fashion for awhile.

Interestingly the Armstrong Whitworth AW.169 to F.155T also included a Q-band radar to provide information on jamming targets out to 8 miles, the main set, (made by GEC, and I assume to be a 33inch AI.18 derivative) had a range of 20 miles. We also know that Elliot Brothers were working on a Q-band seeker for Red Dean though the X band GEC set seems to have been the preference. The UK certainly took the possibility of jamming very seriously.

All in all it is not an impressive story though it would be interesting to know what would have happened had the P.1154RN been pursued, whether the Aspinall CW set would have been ready by 1971/2 or not and whether something interim and AI.23 derived would have been installed. The RAF version was apparently to have a 21inch AI.23 variant so a 30 inch dish version of this is not inconceivable, the Blue Parrot was apparently praised for its ability to detect targets at long range and given that the TSR-2 radar was an evolution of this it seems likely that this may have taken that further. We know that the AI.23 for the lightning went through 4 variants, A/B/C/D. In addition to the 24 inch dish used on the Lightning and the 21 inch dish proposed for the P.1154RAF there would also have been the 18 inch dish for the SR.177 and English Electric proposed a 27 inch dish for the P.8 for the F.155T contest.

If the the capability of the TSR-2 radar is out there somewhere it would probably give us a good indication of what the UK radar industry was capable of in the A2A arena in the 1960s??? It would also be interesting to know what dish size Hawker was proposing for the P.1121?

Whilst scanning the internet I stumbled across the following:

'I was working on the development of a Lightning full authority autopilot to track other aircraft using signals from the Ferranti AI 23b radar'

From: http://www.stevebroadbent.net/253a.pdf

I am particularly poor at technical speak but I take this to mean that the idea was that the pilot could set the autopilot to home in on targets being tracked by the AI-23B??? if so it is an interesting development.

Surely this is something to do with the autopilot running the interception once lockon has been achieved?

Thats mentioned from several sources that Ferranti and EE had actualy developed such a fully automated system.
Date seems a bit late for that though, which I thought was before 1964, though my memory fails me on what dates I have read about that matter.

Zen,

If you remember what sources those are I would love to know. Not much seems to have been written about the AI.23 which is a real shame as it was so central to the UK military aircraft industry for so long!

Scanning the Flight Global archives I came across this:

http://www.flightglobal.com/pdfarchive/view/1962/1962%20-%201911.html

Describes the semi-automatic attack and missile launch modes as used by the Lightning F.3 and then goes on to describe what a fully automatic system would be. The latter corresponds nicely with what Steve Broadbent says he was working on.

Scanning the Flight Global archive looking for something else I noticed that in their cancelled projects list they included the following entry:

Lightning 3 auto-attack system; March 1965; £1.4m

Click to expand...

Basically it looks like the fully automatic interception system was cancelled in 1965 though a relatively large sum seems to have been spent on it.

Source: http://www.flightglobal.com/pdfarchive/view/1967/1967%20-%201672.html

Derek Wood had F.3 auto-attack in Project Cancelled, taken from your Flight find. My very dim memory was this was to be a mod. retrofit to bring F.3 AIRPASS upto then in-build F.6 standard (?AI.23b?)

AI.23 seems to be something of a mystery, it went through 4 iterations- A/B/C/D - but what the differences were I have never managed to establish. The auto-attack idea seems to stem from when the F.3 was regarded as the mythical definitive version to equip all surviving RAF interceptor squadrons. However, it looks like auto-attack never actually happened.

Replying to a necro-thread here, but still likely useful for future readers.

The AI.25 does not appear to be real. I can find only a single mention of it, in Fighters Over the Fleet. The problem is that this places it in 1962. AI.24 dates to the 1980s, so for this to be correct it would seem the RAF planners knew that a new radar would be invented 20 years in the future and they should skip over that number? I strongly suspect that this is a typo that somehow managed to survive until it began being copied by historians. However, Fighters also states the range was on the order of 90 nmi, which places it well beyond the capabilities of any radar of that era that might conceivably fit in the aircraft, and just at the limits of things like the massive ASG-18 of the F-108/F-12, which most definitely would not fit.

As to the AI.23 auto-pilot system, this appears to have been inspired by (and largely copied from) the similar system used with SAGE in the US. The idea was that during the majority of the interception, the pilot's only role was to periodically turn to a new heading sent to them via voice radio. At the same time they were also trying to use the AI radar to gain lock-follow. By sending the angle-of-intercept to the autopilot from the ground, which is where it came from anyway, the pilot could spend all of their time looking at the radar. Once their radar was locked-on, the AIRPASS sight automated the rest of the action, so they simply took the aircraft off autopilot and followed the cue on the HUD. This way, a single-seat aircraft could provide the same level of intercept capability as one with a dedicated radar operator, although this depended entirely on ground-based systems and was thus of considerably less use for long-range aircraft (not a problem with Lightning!) or from forward operating areas (same for RAF).

The AI Mk 24 designation was first used circa 1957, applied to the radar for the OR.329 fighter (cancelled by 1958). On that basis, AI Mk 25 circa 1960 is not impossible.

I've seen AI Mk 25 mentioned in a National Archives file during my P.1121 research. I think it was used for an ambitious CW radar proposal from late 1950s / early 1960s.

The GEC AI.24 from the mid/late 1950s was mentioned in my book:

GEC ‘AI.24’
Not related to the Tornado F.3 radar of the
same designation, this was a GEC proposal
for an advanced OR.329 radar, essentially a
replacement for its earlier X-band 'AI.18 development'.
It was proposed to use the same
Ferranti-designed high-power J-band magnetron
as Ferranti’s ‘AI.23 Development’ with
a 16GHz frequency. The radar antenna was
essentially AI.18 sized (37in swept sphere
scanner with ±60° scan limits or 40-inch
scanner for ±80° scan limits) but range was
initially expected to be at least 30nm for a
75% chance of detection against a 20 square
metre RCS (Canberra, Vulcan) target. GEC’s
experience with AI.18 suggested this might
have been 20% longer in practise and a later
report gave 36nm range for a Mach 0.9
target. The radar weight was expected to be
100lb lower than the X-band ‘AI.18 Development’.
It was expected to be compatible with
the J-Band version of Fairey’s SAGW. In
June 1957 service entry was believed possible
in 1962-1963, though this seems
optimistic.

Click to expand...

AI.24 merely means "the next radar after AI.23" and AI.25 "the radar after the "AI.24". No-one in the late 50s and early 1960s could possibly have imagined it would take until 1980 to introduce a new British radar after AI.23.

overscan said:

I've seen AI Mk 25 mentioned in a National Archives file during my P.1121 research. I think it was used for an ambitious CW radar proposal from late 1950s / early 1960s.

The GEC AI.24 from the mid/late 1950s was mentioned in my book:

GEC ‘AI.24’
Not related to the Tornado F.3 radar of the
same designation, this was a GEC proposal
for an advanced OR.329 radar, essentially a
replacement for its earlier X-band 'AI.18 development'.
It was proposed to use the same
Ferranti-designed high-power J-band magnetron
as Ferranti’s ‘AI.23 Development’ with
a 16GHz frequency. The radar antenna was
essentially AI.18 sized (37in swept sphere
scanner with ±60° scan limits or 40-inch
scanner for ±80° scan limits) but range was
initially expected to be at least 30nm for a
75% chance of detection against a 20 square
metre RCS (Canberra, Vulcan) target. GEC’s
experience with AI.18 suggested this might
have been 20% longer in practise and a later
report gave 36nm range for a Mach 0.9
target. The radar weight was expected to be
100lb lower than the X-band ‘AI.18 Development’.
It was expected to be compatible with
the J-Band version of Fairey’s SAGW. In
June 1957 service entry was believed possible
in 1962-1963, though this seems
optimistic.

Click to expand...

AI.24 merely means "the next radar after AI.23" and AI.25 "the radar after the "AI.24". No-one in the late 50s and early 1960s could possibly have imagined it would take until 1980 to introduce a new British radar after AI.23.

Click to expand...

Impressive and fascinating stuff!

Doesn't this suggest someone thought Fairey was onto something?
Wasn't this the period of bistatic radars and the effort on FMCW, that started so promisingly?
Could AI.25 originally be the backup to the FMCW set, which would be FMICW, making it very ironic.
As then FMICW becomes AI.24.

37" to 40" is definitely F155 territory. The desire to get beyond 40nm was powerfully motivating.
And wasn't a J-band seeker preferred for Red Hebe? Could that preference have translated to the final design prior to cancellation?

Maury Markowitz said:

Replying to a necro-thread here, but still likely useful for future readers.

The AI.25 does not appear to be real. I can find only a single mention of it, in Fighters Over the Fleet. The problem is that this places it in 1962. AI.24 dates to the 1980s, so for this to be correct it would seem the RAF planners knew that a new radar would be invented 20 years in the future and they should skip over that number? I strongly suspect that this is a typo that somehow managed to survive until it began being copied by historians. However, Fighters also states the range was on the order of 90 nmi, which places it well beyond the capabilities of any radar of that era that might conceivably fit in the aircraft, and just at the limits of things like the massive ASG-18 of the F-108/F-12, which most definitely would not fit.

As to the AI.23 auto-pilot system, this appears to have been inspired by (and largely copied from) the similar system used with SAGE in the US. The idea was that during the majority of the interception, the pilot's only role was to periodically turn to a new heading sent to them via voice radio. At the same time they were also trying to use the AI radar to gain lock-follow. By sending the angle-of-intercept to the autopilot from the ground, which is where it came from anyway, the pilot could spend all of their time looking at the radar. Once their radar was locked-on, the AIRPASS sight automated the rest of the action, so they simply took the aircraft off autopilot and followed the cue on the HUD. This way, a single-seat aircraft could provide the same level of intercept capability as one with a dedicated radar operator, although this depended entirely on ground-based systems and was thus of considerably less use for long-range aircraft (not a problem with Lightning!) or from forward operating areas (same for RAF).

Click to expand...

Maury, that's not how SAGE worked. A SAGE-Compatible Interceptor (F-86L, F-89J, F-102, F-101B, F-106) received information calculated by the computers at the Control Center via a digital data link, which would either present a sterring command on the pilot's radar scope, or, if so equipped, steer the interceptor to an Offest Point, which is the optimum location for the attack run to begin - at the offset point, the computer commands a turn to a Lead Collision intercept course, and the Pilot or Radar Observer (On the 2-seat birds) locks on to the target.
At that point, control is taken by the Fire Control System, which can either fly the airplane to the firing point through the autopilot, or by following the commands displayed on the radar scope. The stick trigger is held down as a Consent to Fire, and the weapons are prepped and released by the FCS.
The automatic attack feature was an integral part of all the USAF rocket-armed interceptor F-86D, F-89D, and F-94C and later. I've always been boggled that the AI.23/AIRPASS system didn't have that capability.
Note that the Data Link was used because voice was too slow, required too much bandwidth, and was much more susceptible to jamming. By voice, a controller could handle 1 intercept/minute - with the computer driven data link, a Control Center could handle dozens of intercepts simultaneously.

A good explanation of how the early system worked is here, courtesy of the San Diego Aerospace Museum.


--
Pete Stickney

Question, does anyone have info about the A.I MK17 used in the Javelin and Sea Venom FAW53?
I need the following document if anyone could find it: Radar Type A.I. MK.17, AP N° 2892F