Radar and Fire-Control Automation

[KJ_Lesnick]

I understand why the F4D, F-102, and F-106 had some of the automation they did as they were single-seat aircraft and you'd have a pilot doing all sorts of mental gymnastics to compute the intercept vectors while flying the airplane into the proper position, launching the missiles and breaking away.

Regardless, I'm not sure how much of the automation they built into these aircraft were really necessary.  Sometimes people just take ideas way too far and spend too much money, and end up with issues that affect maintainability and reliability

• Depending on mode, the fire control system could automatically control the plane and maneuver it into the appropriate firing position (F4D/F-102/F-106)
  
• The fire control system could automatically launch the weapons once a predetermined distance is reached (F4D/F-102/F-106)
  
• The MG-10 and MA-1 could use a datalink allowing the plane to be steered into a position at the very least where the aircraft's radar could take over if not carry out the intercept itself, and perform automated ILS landings (F-102/F-106)
  

..
As I see it at least a number of these automation features are very impressive and even cool, but I'm not sure how really necessary they are.

• The technology on the F-102/F-106 and probably the F-6 allowed the airplane to compute intercept vectors which could be displayed to the pilot: This would allow the pilot to simply fly the intercept
  
• The automatic launch capability seems largely unnecessary on the F-102/F-106 even though the AIM-4's didn't have the best range: You'd think a simple "SHOOT" command displayed would be adequate to indicate the plane was in weapons range
  
• The early F4D fired rockets and had a narrower launch window, but I'm not sure if an automated launch was absolutely necessary here either: I could be wrong
  
• The datalink on the MG-10 at least could simply feed displayed data to the pilot and this seemed adequate for allowing the pilot to maneuver himself into position to acquire target with his own radar
  
• The ILS auto-land feature is admittedly useful in terrible visibility
  

I'm thinking the less automation in these systems the aircraft would have required less maintenance, and from what I remember reading about they were horrific at first (particularly the MG-3, MG-10, and MA-1) though they eventually matured into reasonable systems.

What do you think?

[Hobbes]

In those days the pilot's workload was large, huge performance coupled with few electronic aids. ISTR reading comments that the Lightning was bad in this regard. Anything that would reduce that workload in the middle of a fight, meant the pilot could concentrate more on the fight itself, increasing his chances of success.
Early systems may have overshot the mark by being too unreliable, but it was a necessary learning curve and better than not bothering to try.

[Weaver]

With hindsight and in an ideal world, you're probably right. The optimum approach with 1950s technology was to build a big aircraft with a back-seater to drive the radar, lots of speed so it could get into position to intercept the target a long way out, and lots of fuel to stay there for the next one: the F-101B Voodoo approach rather than the F-102. The problem is that this leads to very big, very expensive aircraft that in turns brings affordability into question. Nuclear war is all or nothing: there's no point defending part of your perimeter perfectly if lack of aircraft means that bombers get through another part and nuke your cities anyway.

If you want to defend a big perimeter like North America, then you either have to spend LOADS of money on a useful number of F-108 Rapiers and CF-105 Arrows, or you accept that the planes are going to be smaller, shorter-ranged and single-seat, and then try to deal with the reduced reaction times and high cockpit-workload that implies by throwing technology at it, and remember this was the glorious 1950s when technology was going to solve everything....

Another factor in the high automation of the F-106 generation was probably the desire to ultimately take the pilot out of the plane altogether: the ultimate goal was seen as defence by Blue Envoy-style ultra-long range SAMs. The SAGE datalink system could just as easily be used to provide mid-course guidance to a Bomarc as an F-106. This was partly down to optimistic faith in "space age" technology and partly due to the fact that it was firmly believed in the 1950s that nuclear bombers would be supersonic by the 1960s. A supersonic bomber at 70,000 ft presents such a high closure rate to a supersonic fighter that the mental gymnastics become impossible for the pilot, not least because his small on-board radar doesn't give him anything like enough warning time to do them. He would HAVE to depend on NORAD's big radars and computers to find the target and do the maths, and the margin of error would be so tight that it was better to fire that data straight into his autopilot, or that of an unmanned missile, than give him the chance to screw it up for himself.

[KJ_Lesnick]

Hobbes

In those days the pilot's workload was large, huge performance coupled with few electronic aids. ISTR reading comments that the Lightning was bad in this regard.

I know the Lightning was a nightmare and was too un-automated; however the F4D, F-102, F-106 were too automated.


Weaver

With hindsight and in an ideal world, you're probably right.

I'm not sure how much hindsight was really needed in regards to complexity:  There were complaints in the Korean War about the F-84 and F-86 being too complicated (which lead the drive to the F-104, which seemed ironically more complex than both)

The optimum approach with 1950s technology was to build a big aircraft with a back-seater to drive the radar, lots of speed so it could get into position to intercept the target a long way out, and lots of fuel to stay there for the next one

Correct, divide the workload to two people and you manage to avoid overload.

Admittedly I understand the thinking.  Most fighters at this point were single-engined, and generally some of the best performers were often small and single engined (the exception being the P-38; however the Spitfire, the P-51 Mustang, the P-80/F-80 Shooting Star, the F-86 fit the bill) and most all the early jets were interceptors (they were fast, they could climb quick) except they were generally daytime-use only.

The XF-91 was also another example of a single-engined day interceptor (powered by a J47 which eventually featured an afterburner and four rocket-engines) with supersonic performance intended from the outset.  The USAF specified a radar be installed (same as the F-86D), though it was ultimately cancelled.  Supposedly it could do Mach 1.79 in level-flight (I don't know if this was with rockets and a J47 with afterburner or no afterburner, and whether the radome was fitted or not).  They ultimately cancelled it because it didn't have enough endurance and they had better ideas in mind.

Another factor in the high automation of the F-106 generation was probably the desire to ultimately take the pilot out of the plane altogether

Well, what about the F-102 and F4D?

the ultimate goal was seen as defence by Blue Envoy-style ultra-long range SAMs.

I have little information on the Blue Envoy (like what they intended it to do).  If I recall it's guidance system ended up being used on the Bloodhound but I don't know how the two compared.

The SAGE datalink system could just as easily be used to provide mid-course guidance to a Bomarc as an F-106.

It actually could...

This was partly down to [snip] the fact that it was firmly believed in the 1950s that nuclear bombers would be supersonic by the 1960s. A supersonic bomber at 70,000 ft presents such a high closure rate to a supersonic fighter that the mental gymnastics become impossible for the pilot, not least because his small on-board radar doesn't give him anything like enough warning time to do them.

A Mach 2.5 bomber being approached head on with a Mach 2.5 fighter would yield a closure rate of Mach 5.

While that's very fast, as I understand it, the F-106's radar had a range of 200 nm absolute (this was said on the F-106 site) though it's actual useful range would be less (40-80 nm).  You'd probably have around 65 seconds (Mach 5 is only a mile a second at sea-level at altitude it's less as the speed of sound is slower and a nautical mile is greater than a mile) in theory.  The initial approach would be head-on, followed by a turn to put you abeam the target so you'd actually take longer to reach the plane.

Even if NORAD's computers were feeding data through the datalink (the MG-10 or MG-13) could do this, the pilot would be able to fly himself into position without seeing the target with his own radar.  Once inside the radar's range, the pilot could lock-onto the target using one hand to adjust the radar, the other to fly the maneuver displayed to him and while I assume it'd take a couple seconds to establish a lock, it only takes a fraction of a second to hit the pickle-switch and send a salvo screaming towards them...

I'm not sure about the exact math but it seems to make sense...