Home > Critical Thinking, Current Affairs, Dystopia, Musings, Philosophy sans Sophistry, Reason, Secular Religions, Skepticism, Technology > Reports of Cyberwar against N. Korean Ballistic Missiles are Likely False

Reports of Cyberwar against N. Korean Ballistic Missiles are Likely False

Many of you might have, by now, come across “news” which suggests that the frequent failure of N. Korean ballistic missiles is somehow due to some elaborate “cyberwarfare” by USA. I am sure you must have seen mouth-breathing idiots.. I mean american patriots.. repeat that bullshit because they heard if from some MSM or some ‘alt-media’ shill.. I mean “reliable news sources”. Anyway, the point of this post is to explain why that idea reeks of propaganda and delusion.

But let us be clear about a few things first. It is no secret that N. Korean missiles, either fired by them or in the 1990s by Pakistan, always had a rather high rate of failure. However the reasons behind this rather high rate of failure is immediately obvious to somebody who has read about the general history of developing ballistic missiles and space launch systems. Long story short- it comes down to the choice of fuels.

N. Korean missiles have been traditionally powered by pretty dangerous (but effective) mixtures of old-style hypergolic liquid propellants. Since N. Korean missiles trace their ancestry to Scud missiles, they have traditionally used the same fuel mixture- namely, kerosene and corrosion inhibited red fuming nitric acid (IRFNA) with UDMH aka unsymmetrical dimethylhydrazine as the liquid igniter. Once again, to make a long story short- this particular old-style hypergolic mixture does not scale up well for larger and longer-burning rocket motors

Apart from Russia and to some extent China, nobody has been able to mass produce relatively safe ballistic missiles which use hypergolic fuels of any kind. In fact, the rate of success of early american ICBMS using hypergolic fuels in the 1950s and early 1960s was pretty dismal. While it is possible to build pretty reliable space launch systems using more modern hypergolic fuel combinations, making scores of reliable ballistic missiles which use them requires a lot more experience.

That is why the majority of non-Russian (and now even Russian) ICBMS use solid propellants for their first and frequently also the second stage motors.

Returning back to the subject of ballistic missile control and guidance, let us be clear about a few basics. Firstly, the main guidance systems of such missiles is always internal and almost always based on some form of astro-intertial guidance. In case you are interested about the history of the non-computational side of guidance hardware, here is a link: The Soviet Union and Strategic Missile Guidance. Secondly, the computational part of such systems is quite simple and can be built without using integrated circuits, let alone CPUs.

For example, one of first electronic guidance computers for american ICBMS, known as the D-17B, contained 1,521 transistors, 6,282 diodes, 1,116 capacitors, and 504 resistors. Some of the older Russian designs for flight guidance computers on such missiles even used special rugged vacuum tubes instead of transistors. To put it another way, the flight control and guidance systems of ballistic missiles can be made of very rugged and simple electronic components, especially if you do not require a very high degree of targeting accuracy.

It is basically impossible to remotely “hack” a simple, hard-wired and hard-programmed control and guidance computer in which every discrete component can be repeatedly tested with a multi-meter and oscilloscope.

Furthermore, N. Korea is a pretty paranoid and conservative country. Therefore it is almost certain that they use somewhat primitive but extremely reliable indigenous designs. In any case, they seem to be aiming for targeting accuracy that is between 0.5-1% of distance covered- which is within the reach of such systems. It is therefore my opinion that the frequent malfunctions of longer range N. Korean ballistic missiles are largely due to their inability to scale up an obsolete hypergolic rocket engine technology.

Those problems will however go away once they are successful at building large solid fueled rocket engines. Some of you might know that they have already transitioned away from older hypergolic fuels for their newer short-range (upto 1,000 km) missiles. It is only a matter of time before they do so for their longer-range missiles. If things go the way they are going now, it is possible that they might be able to successfully test and start deploying such missiles in the next 2-5 years.

What do you think? Comments?

  1. P Ray
    April 16, 2017 at 3:55 am

    If things go the way they are going now, it is possible that they might be able to successfully test and start deploying such missiles in the next 2-5 years.

    Right there is the window for the next war. Of course, “for the sake of peace”. Funny how the US doesn’t dare to directly attack their ideological opposites, China and Russia …

  2. P Ray
    April 16, 2017 at 11:45 am

    Talking about “cyberwarfare” against missiles with solid state electronics is sheer stupidity.
    Firstly, the missiles do not have enough memory to support a virus,
    Secondly, the devices don’t even run an operating system that is complex enough to have a virus.
    Plus thirdly, if they use vacuum tubes – they CANNOT be affected by EMP – so in the event of an airburst nuke … North Korea’s nukes could still be launched.

    But as the grasp of electronics escapes most tabloid jorunalists, lots of misinformation gets spread.

  3. A.B. Prosper
    April 16, 2017 at 4:48 pm

    Cyberwar? Nah, not even the people on the Donald Reddit buy that lame propaganda /computer recruiting BS

    In any case North Korea doesn’t network things it doesn’t absolutely have to which is good advice for any country.

  4. April 17, 2017 at 6:02 pm

    on a tangent, this’ll give ya a laugh…

    https://altright.com/2017/04/17/the-doxxing-of-louise-rosealma-reveals-link-between-antifa-and-porn-industry/

    Funny but almost predictable.. especially nowadays.

    • P Ray
      April 17, 2017 at 8:34 pm

      No penetration clips of her, what she’s done so far looks like the nudie clips a girl would share with her boyfriend on snapchat …

    • P Ray
      April 17, 2017 at 8:35 pm

      No penetration clips of her, what she’s done so far looks like the nudie clips a girl would share with her (boy)friend on snapchat.

  5. With the thoughts you'd be thinkin
    April 20, 2017 at 10:58 am

    Heard about Silicon valley’s latet fail, Juicero?

    https://www.bloomberg.com/news/features/2017-04-19/silicon-valley-s-400-juicer-may-be-feeling-the-squeeze

    Its instructions:

    Yes, isn’t it hilarious that so many “high IQ” people from Sill-Valley invested tens of millions in this company.

  6. April 21, 2017 at 10:04 pm

    I think focusing too tightly on hacking the onboard systems lacks imagination. Furthermore, OS’s don’t have to be complex to be hacked. Think about Stuxnet, the virus made the Iranian centrifuges self destruct. The centrifuges were controlled with a simple industrial PLC, spin them too fast and destroy the bearings.


    Where did you copy and paste your response from.. just wondering. Also, stuxnet has virtually no effect on the rate of uranium enrichment by Iran. If anything, it made Iranian design better indigenous centrifuges. Also Iranian MRBMs are working just fine..

    FYI… http://foreignpolicy.com/2017/04/19/the-united-states-isnt-hacking-north-koreas-missile-launches/

    It is believed that many of the parts and raw materials for the missiles are being bought by NK from China. NK may be difficult to penetrate, but Chinese companies are most likely within the capability of the USA to compromise.


    Right.. because N.Koreans don’t understand quality control. You may not like the rest of my response to your reply, but I think it is necessary to get people who think like you out of their self-inflicted delusions of control and power.

    There are many ways to make a missile fail. Hack a CNC program at a supplier/machining shop so that it machines defects into a pressurized tank section so that it bursts. Pull an “in transit” switch of any of the turbomachinery parts with a weaker material, or even a blade machined to the wrong spec so it cavitates when pumping under full RPM’s & pressures. FTS control systems could also be vulnerable, as could GNC. An advanced foe ( like the USA) might layer the compromised systems between boost, staging, terminal guidance etc. Don’t sabotage every flight, mix up the failure modes to make it look more random.

    Do you think that they would use CNC when they have no shortage of skilled indigenous machinists? Do you think even most currently deployed american ICBMs were designed in the era of CNC? Then again, history is probably not your strong point. Also spouting acronyms shows your ignorance, not knowledge. Here is a hint.. the older N. Korean missiles work just fine. Most of their problems occur in missiles which are being rushed into deployment AND use somewhat newer hypergolic propellants.

    Sensors are very vulnerable. Engine mixture ratios could be tricked & you end up melting some critical parts. Spoof the tank pressure system and you burst your fuel/oxidizer tanks on the pad. The failure mode doesn’t even have to be internal to the rocket. A little nightime hanky panky in the plumbing section of the GSE equipment could put foreign debris into the fuel lines, and then that debris gets sucked into a turbopump at 20,000 rpm. Sure you can make a GNC from old style vacuum tubes, but why when off the shelf tools exist. There are so many failure modes in a rocket that a complete fault tree analysis or FMEA would be over 1000 pages. An advanced enemy like the USA has the resources to exploit many different paths. It is not unreasonable to think it wouldn’t be worth a try.

    Perhaps jerking off to technical jargon which you do not understand is your hobby.. well to each his own. Maybe you, like many others in USA, you actually believe your own bullshit. Here is what you might not know. They make pretty much all their weapons and components inside their own country- one which western intelligence has basically zero access to. Also, they are more paranoid about this stuff than you can imagine.. can imagine.

    You may well be right that there was no sabotage, it would be effective enough to get inside the heads of the NK leadership, skullfuck their scientists, and make them all paranoid in their missile development work. This could at least slow them down & make them work extra hard on testing and sourcing more systems on their own. The bad part of this approach is that once they learn with certinty how to do it all on their own, they will have a more robust capability.


    Are you really delusional enough to think that N.Koreans are that stupid? You do realize that they are not white american working in silly-valley.. right? They are far smarter and tougher opponents than you realize. Perhaps the scariest outcome of this is going to be them just going for solid propellant based missiles with some liquid propellants in 2nd or 3rd stages. Interestingly, that is the solution which ICBM programs of every other countries have inevitably lead to date.

  1. April 16, 2017 at 2:28 pm

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: