4
$\begingroup$

As we know, during the Space Race, the United States and the Soviet Union were in fierce competition. Each trying to best the other by creating their idea of a moon rocket. While the United States's aprroch was to increase the size of their engines(resulting in the F-1 engine), the Soviet Union simply decided to add more engine to each stage.

Each design choice at first may seem equal, however, the records of both rockets tell otherwise. While the succes to failure rate of the Saturn V was nearly perfect, the N-1 failed on every one of its four launches.

While it may be tempting to simply say the Soviet's rocket design was dumb, it seems that many other prominent factors played a role in the failure of the N-1 program. Such as this excerpt from Wikipedia states:

N1-L3 was underfunded and rushed, starting development in October 1965, almost four years after the Saturn V. The project was badly derailed by and the death of its chief designer Sergei Korolev in 1966.

Along with above, there seemed to be strife and disagreement between certain, key project leaders, which also probably caused friction in the project. Furthermore, from what I've read, the engines used on the N-1 were fairly well designed and tested.

In short, it seems that the failure of the N-1 rocket was a result of many external problems, instead of it simply having a "dumb" design.

This brings me to my question. If The problems above were eliminated, and a N-1 rocket design was given proper funding, R & D, etc., then could a working N-1 rocket(or a rocket of similar design)be successfully constructed and flown today? Or does the N-1 simply have a bad design?


NOTE: my use of the word "modern" in the title is a little different from its actual meaning. When I say "a modern N-1 rocket", I simply mean a rocket which would be designed, built, and tested today, without all of its past design problems.

$\endgroup$
  • $\begingroup$ Your question is a little underconstrained. Do you allow infinite budget and time? $\endgroup$ – Organic Marble Mar 19 '17 at 1:46
  • $\begingroup$ I'm afraid even given proper budget, the answer to this question may be impossible without actually attempting to build the N-1. Reason: It did have design flaws; probably originating from its troubled design history. Whether they could be "just fixed" being circumstantial details, or are ingrained in the core design, is impossible to tell unless we know them and the steps to fixing them - essentially, until we have complete blueprints of a flawless N-1 (or a clear proof that the design can't be fixed, and why.) In short, the only way to answer this, is to build a working N-1. $\endgroup$ – SF. Mar 19 '17 at 2:34
  • $\begingroup$ @OrganicMarble My apologizes. I see how you could be confused. I'm a little confused though, as to how exactly answer your question. Not for lack of knowledge, but because I'm not sure how yo correctly explain. Think of it this way: Theorectially, would the design of the N-1 work? $\endgroup$ – Christian Dean Mar 19 '17 at 2:51
  • $\begingroup$ @SF. Really? That's a bit disappointing. I didn't think it'd be that hard to explain if a design would work or not. If you really this question is impossible to answer, I'll delete. $\endgroup$ – Christian Dean Mar 19 '17 at 2:54
  • $\begingroup$ There's a continuum between "launching the N1 as it was designed", which we know doesn't work, and "redesigning the N1 little by little until it's a Saturn V", which we know would work. I believe you don't have to go all that far along that continuum to get a working rocket. $\endgroup$ – Russell Borogove Mar 19 '17 at 3:25
13
$\begingroup$

According to Wikipedia, the developmental problems with the N-1 were exacerbated by the fact that the NK-15 engines

had a number of valves that were activated by pyrotechnics rather than hydraulic or mechanical means, this being a weight-saving measure. Once shut, the valves could not be re-opened.

This meant that individual engines could only be tested by flying them on the N-1. After the first failed flights, while the rocket was being redesigned, the NK-15 was replaced with the NK-33/43, which has since proven to be a successful engine, so we can assume that problems with individual engines could be worked out. Longitudinal "pogo" oscillations of the engines tore plumbing in the test flights, causing various fires and explosions. Anti-pogo measures could be installed, and sensors could be better-calibrated to shut down vibrating engines before they tore something loose.

The article also mentions

exhaust plume and fluid dynamic problems (causing vehicle roll, vacuum cavitation, and other problems)

It's hard to describe how much more advanced the state of the art in software simulation of fluid dynamics is now than it was in 1969. A typical desktop computer today is something like 1000 times more powerful than the fastest supercomputers of that era. It's very likely that these problems (presumably due to the interaction of the exhaust plumes of 30 engines with the airflow around the body of the rocket) could be identified, but solving them might require extensive redesign of the first-stage thrust structure -- for example, it might turn out that one or two engines at the centerline of the rocket would solve the problems created by the N-1's concentric-ring engine arrangement. Whatever the cause and potential solution of these problems are, they're probably the most "fundamental" of the N1's design faults.

Roll control on the N1 was through grid fins, as the first stage engines weren't gimbaled; pitch and yaw control was done by differentially throttling the engines. Aerojet has since added gimbals to their version of the NK-33, which would give far better roll control at the price of additional weight and complexity.

A secondary source of problems in the test flights was the KORD engine control system. One of KORD's jobs was, on failure of a single engine, to shut down the opposing engine to maintain balanced thrust. On the first flight, transient signals from pogo-induced damage caused KORD to shut down the entire stage -- not ideal, but the stage was on fire by that time anyway, so it's likely that the rocket would have been lost regardless. More testing, better quality control on sensor installation, and so forth could probably mitigate these problems as well.

Because all 4 of the N1 test flights failed during the first stage portion of flight, none of the upper stages were ever tested in realistic conditions, and doubtless they would have had their own problems (though those problems would not include surviving ten million pounds of thrust). I don't imagine that any such problems would be insurmountable.

I think the N1 ultimately failed because not enough time and money was spent, and because the people who could see the engineering problems didn't have the power to set the schedules. In contrast, development of the Saturn was very successful, with an almost flawless first test flight (Apollo 4), and a very good record: I believe 12 of 13 successful launches and no catastrophic failures. Allowing Apollo 4 to be delayed by a full year from its original launch schedule made that possible. As Feynman said in regard to the Challenger disaster, "for a successful technology, reality must take precedence over public relations, for Nature cannot be fooled."

$\endgroup$
  • 1
    $\begingroup$ Feynman, as quoted, seems grotesquely inconsequential, as PR also belong to "reality" and "Nature". Bad PR can demonstrably fail a technological project by cutting its budget. $\endgroup$ – kubanczyk Mar 19 '17 at 12:08
  • 3
    $\begingroup$ PR (and its precedence over the reality know from science and engineering) is what doomed the Challenger (by demanding a launch outside of known-safe launch parameters). There's nothing inconsequential about that. $\endgroup$ – Hobbes Mar 19 '17 at 19:14
7
$\begingroup$

It's worth noting that the N-1 launch failures were not unexpected. Early on in the program, a decision was made not to invest in facilities for ground-testing complete stages, because building the first-stage test stand would take 4 years and a lot of money.

Instead, the plan was to do 12 test launches, all using boilerplate payloads. This was a common approach in Soviet rocket projects at the time. The first 2 launches were done in quick succession (February and July 1969), but after Apollo 11, the program lost its urgency as the Moon race was lost. Work continued at a slower pace, with the fourth launch occurring in 1972.

In 1974, Mishin's OBK-1 (in charge of the N-1, and still working on a Moon mission and improvements to the N-1) was combined with Glushko's OKB-456 under Glushko's leadership. He had opposed the N-1 from the beginning and immediately cancelled it. This happened in May 1974.

This termination came at an unfortunate time: the fifth rocket, with restartable NK-33 engines, was being readied for launch, and more rockets were in the pipeline. The fourth launch had come close to doing a full-duration burn of the first stage, so they were hoping to be able to test the second stage on the fifth flight.

(source: N-1 for the Moon and Mars, N. Stevens and M. Johnson)

$\endgroup$
3
$\begingroup$

The development of the F-1 rocket engine for the first stage of the Saturn V took a long time and was started years before the development of the Saturn. In 1955 the first studies for the engine were started. In 1958 the first contract for the development was given. The first test of a combustion chamber was 1959, this test was done without the turbine and turbo pumps. Special high pressure tanks were build for this ground test to use only pressure feeding. 1961, before Gagarins flight and Kennedy's speech the combustion chamber was successfully tested. 2471 tests were done with the F-1 and 1110 tests for the full time needed for a flight of the Saurn V. There were 34 tests of a complete S-IC. 1966 the engine was certified for manned flight.

Given enough time for development and tests of the N1 and its engines as well as enough money for numerous ground tests of the engine itself and a complete first stage, a modern rocket like the N1 should be possible. But the interaction of all 30 engines was very complex and difficult.

$\endgroup$

Not the answer you're looking for? Browse other questions tagged or ask your own question.