The ISS had problems related to ammonia more than once.

These problems happened in different calender years, and it seems like almost every "near miss" that makes the news about the ISS is connected to the ammonia loop. This system exists to transfer heat from the station to the radiators, which I've heard compared to a car radiator.

When one single subsystem consistently puts the entire station at risk, I would predict that NASA makes it a high priority for technology improvement. Particularly because they've actively been talking about BEO space stations quite seriously (which I'm sure would multiply the complications).

Are there any competing technologies for the cooling systems of the next generation of NASA or commercial space stations? Hopefully options less likely to leak... Or is it generally accepted that we'll still use the ammonia coolant in spite of its track record?

I don't mean the distant future either, I want to limit this to concepts that a mission designer could currently take seriously.

  • $\begingroup$ Are you against the concrete cooling fluid ($NH_4$) or the idea of using radiators and heat exchangers in general? $\endgroup$ – Deer Hunter Jan 14 '15 at 19:26
  • 1
    $\begingroup$ @DeerHunter I'm not against either of those, but if you design a new radiator using the same cooling fluid, then how would that impact the multitude of failure modes the ISS saw? Sometimes incremental improvement is the best option. So if there is some fantastic redesign of NH4 radiator loops, then I would still call that a next-generation cooling technology. $\endgroup$ – AlanSE Jan 14 '15 at 19:49
  • 2
    $\begingroup$ Some technologies to be evaluated by Nasa. $\endgroup$ – mins Jan 15 '15 at 6:26
  • $\begingroup$ Shuttle used Freon in its external cooling loops. I don't know why they switched to ammonia for ISS. $\endgroup$ – Organic Marble Aug 31 '15 at 13:01
  • $\begingroup$ Work on a morphing radiator is kinda under way. $\endgroup$ – Deer Hunter Nov 5 '15 at 11:57

There are multiple possible improvements for moving the coolant, in pumping systems. There are piezoelectric pump and electrohydrodynamic pumping systems being worked on, with main advantage being increased reliability. One other candidate would be thermoelectric aka Peltier effect pumps. Similar tech would have ample applications in terrestrial cooling solutions.

You can find state of the art and proposed future solutions discussion in NASA technology roadmaps, for this it is the technology area 14, thermal management systems.

There have been multiple different issues with the external ammonia cooling loops, from what I understand. Two that I know of was a pump failure, that had several contributing causes. Pump was returned to earth and taken apart. Basically operating conditions pushed several components (bearings, seals) of the mechanical pump outside of the design margins, resulting in wear. There are also issues with valve failures inside PVTCS, which I don't think has been fully root-caused. In each instance, contributing causes have been mechanical complex systems that are quite hard to get right on orbit, even though electronics failures have been suspected.

At a very high level, the possible mitigations are iterating and improving the mechanical subsystems, by understanding the operating conditions better one can design better margins, use better materials and subassembly designs etc.

The other possible mitigation is to try and minimize moving parts as much as possible and use solid state systems where possible. That's where the future currently TRL 3-4 level technologies come in, mentioned above.

There have been a plethora of operational lessons as well of course. Ability to return failed subsystems to earth for quick analysis and design improvements is a key one, as the best way to improve technology is still to operate it and learn from the process.

Some references:






  • $\begingroup$ Instead of pointing to a link, could you elaborate on the future solutions here? $\endgroup$ – Vedant Chandra Nov 5 '15 at 18:48
  • $\begingroup$ A v.solid answer. $\endgroup$ – Deer Hunter Nov 5 '15 at 20:15

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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