Thread by @Indrani1_Roy

For me, NGLV is the most exciting developments from ISRO. Enough info has trickled out to make some educational guesses.

Thank you @TitaniumSV5 for capturing the slide Dr. V. Narayanan, Director, LPSC.I had some first-order questions which are now answered.

Beware, long thread.

https://twitter.com/VikranthJonna/status/1750184779210002644

@TitaniumSV5 In this thread I will answer 3 basic questions:
1. Why does the first LM470 use 9 liquid methane-oxygen (methalox) engines?
2. Why develop new methalox engines, when ISRO is already developing semi-cryo kerosene-oxygen (keralox) engines for LVM 3?
3. Why 3 stages?

@TitaniumSV5 Q1. Why does the first stage of NGLV is going to have 9 engines.

Ans: The reason is actually quite generic. This magic number of 9 is common to a lot of modern launchers designed to:
1. Have a reusable booster, and /or
2. Have the same engine in the first and second stage.

@TitaniumSV5 Answer1: The mass-fraction of fueled_mass:empty_mass of the first stage is roughly 10:1.
The LM470 booster will carry 470 tons of fuel & will prob. weigh ~60 tons empty. I'm using a higher empty weight as ISRO is not using the most optimal design. More on this later. Contd...

@TitaniumSV5 When it comes in to land. The thrust of the engine should be around 60 tons (or around 600 kN) too. Otherwise, it will start going up again.

The thrust of each methalox engine is around 1160 kN and it can be throttled down to around 600 kNs too!

Contd…

@TitaniumSV5 Also, the difference in the thrust between the 1st & 2nd stages is also ~9:1. So they can design one engine that can be used for both the stages with some adaptations, e.g. nozzle etc.

This allows for lower development cost & manufacturing costs (better "economy of scale").

@TitaniumSV5 Also 9 engines allows the one engine which relights during landing to be placed in the center and the other 8 engines to be symmetrically placed around it.

This gives a lot of redundancy during the ascent phase as well should an engine fail!

@TitaniumSV5 Now, we come Q2.

Why is ISRO developing new liquid methalox engines when ISRO is already developing the semi cryogenic SCE200 engine?

Here the word "new" is misleading. While chronologically correct, the LM100 is closer to ISRO's proven liquid workhorse engines.

@TitaniumSV5 SCE200 is a staged-combustion engine. These engines are the most efficient and the most complex.

ISRO's only experience with this class of engine is CE7.5 from Mk2.

Please focus on the Isp (efficiency) of CE 7.5 engine on the left and CE20 engine from LVM3 on the right.

@TitaniumSV5 On top of this limited experience and success with the staged-combustion engines, ISRO is going for 3-times the chamber pressure on the SCE-200 engines than any of its previous engines.

@TitaniumSV5 On the other hand, LM100 is a gas-generation engine. ISRO has a lot more exp. and success with GG engines. Not just CE20, but Vikas engines also use GG.

There are Vikas engines optimized for sea-level, vaccum. They have also tested deep throttling.

@TitaniumSV5 In fact, ISRO has already had success with a successful realization and testing of a subscale LM10 engine whereas it is facing some challenges with the testing of the SCE200 engines.

But as a stage, SC2000 stage is further along.

@TitaniumSV5 Another aspect is the choice of fuel. For the first stage, we will discuss the difference between kerosene and methane.

For the 3rd stage we will compare the liquid hydrogen and methane.

@TitaniumSV5 Kerosene & methane produce interesting tradeoffs.

Methalox engines are more efficient. But, Kerosene is denser, thus requiring smaller tanks, thus lighter stage, aka a more efficient stage!

Methane is cheaper to acquire. Kerosene is cheaper and easier to transport and store!

@TitaniumSV5 In the case of an expendable launcher, it’s a pretty even toss-up. But for a reusable launcher, methane takes the cake, because methane burns more cleaner leaving behind less residue (coke). Hence it is preferred.

@TitaniumSV5 Now, we come to the question which is most unique to NGLV among modern launchers: Why does it have 3 stages where 2 seems to be the most prevalent?

I am also sad to say that this is where I have read the most confusion and reasoning on Twitter/X.

@TitaniumSV5 Obviously, this step is towards 2 goals. The increased complexity will yield:

1. better payload fraction than a 2-stage rocket.
2. higher flexibility for launch to LEO and GTO.

One is obvious, let's understand 2 a little better.

@TitaniumSV5 Rockets that are optimized for launch to LEO and GTO are architected differently. LEO is a circular orbit roughly ~500 km above sea-level. LEO satellites are injected at relative velocity of ~7.5 km/sec. GTO is highly elliptical orbit where satellites are injected at ~10.2 km/sec

@TitaniumSV5 The problem in using rockets is that the majority of the energy of the rocket is expended in imparting the velocity to the rocket itself, rather than the payload!

Therefore, it makes inherent sense to make the last stage as small as possible for the last stage for GTO injections

@TitaniumSV5 Let's dig a little deeper.

Although the workload distribution between the stages for LEO/GTO optimized rockets is different, the job of the first stage is the same.

Get the hell out of the atmosphere as soon as possible.

@TitaniumSV5 Atmosphere adds drag and also reduces efficiency of rocket engines considerably.

This is why people use solid-boosters. Although they have the lowest efficiency, they burn fast and create a lot of thrust allowing the rocket to escape the atmosphere quickly.

@TitaniumSV5 Once out of atmosphere, the optimizations between the two architectures come to bear.

For geo-stationary, there are much more efficient ways to "raise" the orbit to circular 36,000x36,000 km (called Hohmann transfer). The focus is to raise the delta-v.

@TitaniumSV5 Here, comes the first trade-off. The most efficient cryo-engine known to man uses liquid H2.

The problem is that liquid H2 is also the least dense, and the tanks (and therefore the last stage) required becomes large and heavy.

Compare 2nd stage of Ariane5 & Falcon9.

@TitaniumSV5 So for H2-stages, the Isp part in the rocket equation is increased and the mass fraction (m0/mf) part is reduced.

However, since the delta-v is directly proportional to ISP and only logarithmically proportional to m0/mf, hydrolox engines are preferred for the last stage.

@TitaniumSV5 In general, the goal is to make the smallest and most efficient last stage.

This is where the CUS stage of GSLV Mk2 shines. There's never been a more efficient engine developed by ISRO to-date. There's none on the horizon either. And that's okay.

@TitaniumSV5 Finally, there is another problem with using a hydrolox stage.

The faster a rocket imparts the delta-v, the lower is the cost it pays in terms of "gravity-drag" (I will let you Google it).

But hydrolox engines typically have the lowest thrust of all cryo-engines!

@TitaniumSV5 To lower the gravity-drag, the trajectory is made as flat as possible (the more perpendicular your thrust is to gravity, the lower is gravity drag).

Compare, LVM-3's launch trajectory for GTO vs LEO. For GTO, the last stage barely climbs 40 kms while imparting 60% of the delta-v

@TitaniumSV5 On the other hand, the last stage must do all the work and "climb" all the way to the LEO orbit. For On-web launch, gains around 300 kms!

Again, the rocket equation comes to play. Methalox/kerolox engines are less efficient, but the stages are shorter and lighter than hydrolox.

@TitaniumSV5 What tilts the choice to methalox and kerolox engines is that they can impart the delta v much faster, hence paying the least gravity-drag.

If you have a methalox lower stage, you might as well use the same for the last stage. Same for kerolox.

@TitaniumSV5 To put everything in perspective, the keen reader is pointed to Ariane 5 as a GTO-optimized architecture and Falcon 9 as a LEO-optimized architecture.

Ariane5 could lift 20 Tons and 10.85 Tons to LEO and GTO respectively. For F9, these numbers are 22 tons and 8.5 tons.

@TitaniumSV5 ISRO is trying to address this trade-off by introducing the 3rd-stage.

The larger methalox LM70 stage with twice the fuel and much larger engine is optimized for LEO-launches. But for launches to GTO, the smaller, lighter and more efficient C-32 stage will be preferred.

A couple of final notes. Finally, I am seeing somethings that I have been eagerly waiting on GSLV Mk2 and LVM3 to finally get addressed in NGLV.

The corrugated interstages are likely going to be replaced by isogrids.

Also, FINALLY ISRO is embarking on the common-bulkhead for its cryo-stages. It was discussed for C-32 stage. It would have increased GTO payload by 590 kgs! But that never materialized.

With this LVM3 with SC120 could have lifted 5.7 tons to GTO!

But, now they are showing that LM70 will use a common-bulkhead.

It is easier to have a common bulkhead for the methalox stage with the difference in boiling temperature of liq. CH4 and O2 being 20C (rather than 57C between H2 and 02)

I hope, that this bulkhead is progressively adopted in the C-32 stage, and also the lower LM120 and LM470 stages.

What ISRO does with what it 'has' is amazing!

NGLV will be a real augmentation to what it 'has'!