Some of this some of you know and I also apologize in advance for any repetition. I am putting some of my thoughts, research and views together in this.
In the mid-1980s the first weapons were assembled. Sundarji knew of them. By 1987-88 (Brasstacks time) there was the first move to mount weapons on aircraft. The Mig-23/-27 and Jaguar were all considered.
The weight of the device was limited by the ability of the platform's centre pylon to carry a certain mass and the ground clearance.
The weight of the nuclear weapon selected - arming mechanisms, casings, fins, fuze/ retardation configuration included - selected came in at about 450-550kg (perhaps less) but with minimal ground clearance - lack of coordination between DRDO/BARC and the IAF being responsible. Revisions were of course made.
Though by 1993-94 the Jaguar had been superceeded by the Mirage 2000, for years it was the primary option.
At all stages the IAF knew the following:
(a) The number of weapons
(b) Their yield
(c) Their weight.
Of the following questions, I asked each and every one of them to an IAF Air Marshal, now retired who answered in the affirmative. The IAF knew a great deal more than the Indian army about India's weaponization in the 1980s and into the 1990s.
At that time yields of 15kT were the ones spoken of, but realistically 25KT is very much possible. Note however, that all the discussions I had referred specifically to operational 15KT weapons. I don't think a quibble about 15-25KT is going to be a big deal.
George Perkovich, writing in India's Nuclear Bomb at page 242 describes a 1982-83 upgrade of the POK-1 device weighing 170-200kg and also mentions a FBF design being readied for testing at this time.
This 170-200kg design is very important as it dovetails in with a statement made by Karnad at page 82 of his book India's Nuclear Policy where he states:
"India also has the 300km Brahmos supersonic cruise missile (SCM), produced with Russian collaboration and can deliver a conventional warhead. A 20KT fission weapon is being specifically designed for it by BARC..."
This would tie-in with my consistent comments that 700-1000kg is excessive for a 20-25KT Indian fission warhead. Given that the Brahmos payload is 300kg, Perkovich may well be correct.
To add to the story, in 1996-98, prior to the tests but around the time of the Su-30MKI purchase, the MKI was selected in part because its centreline pylon was to be used at some point for a stand-off nuclear cruise missile.
Note where the air-launched Brahmos is ending up at present ?
Hammerblow IV was an IAF exercise in which a nuclear backdrop was factored in and the IAF demonstrated toss-bombing techniques: note that such techniques are not unique to nuclear strike and the MiG-23BNs involved were aiming at a heavily defended installation. The technique, however, it not an easy one to master so training pilots to do it does show an ability to delivery nuclear weapons in that way (the aim being to escape the blast).
During the period 1993 to 1999, the missile deterrent existed largely in potentia.
The Prithvi entered low-level production in 1993 with a rate as low as 10 per year but has now reached a total of between 180 and 200.
The Agni-TD was "operationalised" after a fashion with a few of them being around "just in case". None of these weapons were transfered to the armed forces.
The 1999 test of the Agni-2 was the first step towards a missile based deterrent. During the Kargil war, however, though 1-3 Agni-2 prototypes and perhaps 4 Prithvis were operationalised as a missile based deterrent in the event of a Pakistani strike. This was essentially a DRDO/BARC contingency force rather than a fully operational deterrent.
In the period 1998-99, however, Defense News reported that some 20 Agni missiles existed. I think that highly unlikely. The figure for Agni-2 was at best 3 and Agni-TD, 4.
I will not go into the story of the slow deployment and production of the Agni-1 and Agni-2.
Back to the central nuclear weapons story.
In 1996 when I started probing into weapons capability, BARC was very clear that it saw itself being able to produce fission or boosted-fission weapons if needed without the need for a dynamic test.
In 1996, despite Rao's moves in 1995, BARC was convinced that no Government would give a go-ahead for the dynamic testing of anything.
In 1998, what were tested - sub-kts aside (which were intended for simulations and probably to assess the use of reactor-grade plutonium) - were intended to - according to my sources:
(a) Validate the stockpiled weapons
(b) To develop a core design for a TN device while validating FBF technology which could be stretched to produce a variety of weapon designs without recourse to another test.
At all times, BARC wanted a series of 9 tests but realised that there were going to be political ramifications that would not allow another round of tests.
I have no real indication as to what S6 was so I will not speculate. Please note that the designations of S-1 etc are what we are using and have no official validation.
Regarding the weight of S-1, Raj, please see my reference to Test Simon of 1953 in Nevada to see an 11000lb TN 43kt test. The device was designed as a lead in to a much larger weapon - I think that would account for its weight. Similarly S1's 450kg should be considered in that light. The actual weapon designed to emerge may be lighter, heavier or something similar.
I am not going to debate yields. I agree with Iyengar's position that more tests are necessary and I am in absolute agreement with Srinivasan's analysis in respect of India's fission, boosted-fission and TN capability:
(per page 68 of Karnad's nuclear policy)
"According to Srinivasan,
without more tests, the reliability of the 20kt fission weapon is "100%" of the tritium boosted design -"It will surely work... But the full performance of the booster part will be subject to less than 100% surety; [and] in the case of the full thermonuclear device the confidence level will perhaps be less.""
2002 was the watershed year for the military and nuclear weapons.
The Agni-1 had been tested, Agni-2 had entered LSP and the army - angling for control of these weapons - asked a direct question of BARC and DRDO regarding the status of the deterrent. Over a period of 48 hrs. Gen. Padmanabhan came to understand concerns over the TN weapon. He was told that BARC could not give a 100% yield assurance but that the TN capability was there.
This was also the period when DRDO and BARC began getting their own act together:
Quoting from Karnad's India's Nuclear Policy:
(at page 82) "Until 2002-2003 the high firewalls between the nuclear weapons design program and ASL prevented the optimization of nuclear missile design. Now there is very close cooperation. Thus, for example, S.K. Sikka, as head of the thermonuclear weapons project was asked to change the size parameters of a weapon for a certain Agni missile by a mere 5cm to exactly fit the missile configuration -because changing the missile size would require massive redesign and development work resulting in a delay of 3-4 years. After some quick calculations, Sikka readily agreed to chance the dimensions of the weapon in question. This level of cooperation is a great improvement on what existed when R. Chidambaram headed AEC."
At that point, there was a very conscious move not to rely on TN technology and to actively rely on the fission and boosted-fission technology available.
Scalability of these weapons is not easy, but it can be done and the devices so made would be more reliable than those on TN technology at this stage.
What is India's capability in this regard ? Fission technology - if one is willing to pay a weight and fissile material penalty - can be realistically scaled to 60-80KT in the Indian context with weapons immune from pre-detonation problems. Anything larger than that - problems can emerge (the 120kT French MR31 and the monstrous 500kT US Mk.18).
Would you want to test this again ? Sure. But the core fission technology has been mastered.
As far as FBF weapons are concerned, the largest one deployed gave a yield of 500kT.
Can India scale the FBF primary of S1 (assuming it was 17kT and worked properly) to 500KT ? No.
Can India scale such a weapon to 100-200kT ? Yes.
Would such a weapon be 100% reliable without testing ? No.
It would also be very bulky and weigh between 750-1000kg - you notice how this figure keeps recurring (Agni-2 payload etc) ?
See Srinivasan's comments which I posted earlier.
With FBF, something will work, but the efficacy of the boosting will at this stage be subject to less than 100% surety.
Can Indian scale its FBF to 60-80kT ? Yes - with probably a greater degree of reliablity and certainly enough to be acceptable.
With respect to the TN design - without going into the yields - the question comes down to this:
Whatever happened with S-1, can BARC weaponize a TN weapon that will work without further testing ?
The answer I keep getting is "probably".
Have they done it ? "Possibly".
I also note Chidambaram's words to the effect that BARC has been continuously working on things over the last 11 years. To assume things are static is dangerous. Indeed, Karnad at page 71 of India's Nuclear Policy states that:
"There are over 12 untested weapons designs, encompassing nuclear and thermonuclear gravity and glide bomb and warheads optimize for delivery by aircraft, land based cruise and ballistic missiles and ship and submarine-launched cruise and ballistic missiles on the shelf."
The most conservative of these designs could be viable with out dynamic testing - the most ambitious would be a gamble.
Regarding TN weaponization, I would say "weaponization" is a strangely undefined word.
Does it mean a fabricated device that is ready to be either tested or mated to a delivery system ?
Or does it mean a fabricated device that is ready to be handed over to the user with the designers giving an assurance of its viability ?
If it is the latter, I do not believe it has been done for any Indian TN device. The former, probably yes.
Where does this magic 200kT figure come from ?
It was intended that the S-1 device provide the basis for a weapon up to a "maximum-minimium" 200kT yield.
It was also intended that a contingency 150-200kT FBF be prepared in the event of the TN not working to specifications.
Does India have the capability to make a 200KT TN weapon ? Yes.
Has it made such a weapon ? Probably.
Is it 100% certain it will work to specifications ? No.
Is the SFC satisfied with that ? Not yet.
Is it deployed ? No.
Does India have the capability to make a 200kT FBF weapon ? Yes
Has it done so ? Probably
Is it 100% certain that it will work to specifications ? No.
Is the SFC satisfied ? More than with the TN
Is it deployed ? Probably not yet.
What we do know is that the Agnis are being deployed with payloads between 700 and 1000kg.
We also know that whatever is being said, the 20KT fission weapon - if it can fit on Brahmos - weighs under 300kg.
Therefore whatever is carried on the Agnis is over 20KT and weighs between 700 and 1000kg (say between 500kg and 750kg to cater for decoys etc.).
India's SFC has tailored its requirements and its planning around what is reliably available.
Such a capability is known and understood by the SFC and the GOI - hence Adm. Mehta's statement which is very conservative:
"We are a nation which maintains a credible deterrent...more than enough to deter anybody,'' said Admiral Mehta. And should someone do the unthinkable by launching a first-strike, then the "consequences will be more than what they can bear...
As far as we are concerned, scientists ... tested[/b].''
Air Marshal Ajit Bhavnani's words to Karnad also speak of greater synergy between the political establishment and the SFC(page 99 of India's Nuclear Policy):
"It has been time consuming for the political bosses to understand what's a de-mate situation, what's a mated situation, why we should have a mated situation and when a de-mated situation...But once they were made to understand, we are now in a good situation."