ISS Cooling Problem

Peripherally impacted by the spacewalk repairs, I started looking for more technical information and found a detailed description:
http://www.nasa.gov/pdf/162182main_STS-116_Press_Kit.pdf

However, this doesn't do a good job of explaining the big picture:

• "A" loop - this is an ammonia cooling loop that uses the external radiators to dump heat. Ammonia is the coolant loop gas, not freon, that is compressed, condenses to release heat, and the liquid pumped to a heat exchanger to cool by evaporating. But ammonia is toxic so no part of that loop enters the human occupied space. FYI, freon is less toxic BUT in the limited atmosphere of a spacecraft, not something you want displacing the breathable air.
• "B" loop - this is a water, heat exchange loop that is chilled by the ammonia system and then 'cold plate' cools the electronics and interior.

The problem is unregulated, the ammonia loop can freeze the water. This would crack the heat exchanger and risk ammonia coming into the living space, the interior of the station.

The primary, ammonia control valve, a proportional valve, has failed, apparently in the "cold" mode. But the controllers have found they can operate an ammonia isolation valve, normally and ON/OFF to regulate the cooling . . . sort of. They are commanding it to turn ON or OFF and then 'yanking the power' to leave it in a partial state. In effect the ON/OFF valve is providing proportional control with the ground controllers 'in the loop.' But there is a problem that to control the valve accurately, they really need 0.1 second control and humans and control loops at best are in the 0.2-0.3 second range.

Now hindsight is always 20/20 and obviously one wonders if having two proportional valves wouldn't have worked better. The second valve replacing the ON/OFF isolation valve. Then if the pump assembly has to be replaced, command the working valve to OFF and problem solved. Alternatively if the ON/OFF valve can stop at a proportional setting by pulling the power at the right time, use a simple timing circuit with two such valves and eliminate the proportional valve. Regardless, the clever trick is figuring out to sustain station operation by using the ON/OFF valve as a pseudo, proportional control valve.

My impact is each 'space walk' results in a 'mission freeze' on the network I work on. Even though my small part is data recording, in an excess of caution, my maintenance activity scheduled two weeks earlier, Dec 23 through Jan 6, is at risk of being canceled. The activity is to fix a problem that leads to rebooting our monitoring systems due to a latent Oracle vs. Redhat bug. But this has always been a risk associated with maintaining a NASA network.

Hindsight says I should have scheduled a series of back-to-back, daily maintenance activities, one for each of my 14 systems. If some unplanned "mission freeze" occurs, simply cancel the impacted daily upgrade and then double-up on one of the other activities. <grins>

Assuming my big activity is going to get canceled on Monday, I'll smile and say, "Thank you Sir! May I have another?" We have had no luck negotiating a modification to an activity. So when the activity is killed, I'll smile and schedule 14, one-day, activities for each of my systems. Meanwhile, I'll preload the upgrade software on the systems.

If a system fails before the upgrade, I'll 'make operable' by doing the software upgrade at that time and then close the ticket. There is no fault nor foul if an activity is canceled by the requester before the window. So with 14, one-day activities, the probability of a mission problem blocking my activity is substantially reduced.

In reality, all 14 systems could be updated in one activity but 'putting all eggs in one activity basket' is what is about to delay my current activity.

Bob Wilson

 

Actually I would not have cared but it looks like my activity is going to be canceled on Monday. So I'll do plan "B" . . . a series of 1 day 'scheduled activities,' perhaps with an idle day in between, and do the updates then. I'll schedule one activity per system but 'opps' I'll do as many in parallel as I can in the first ones. Then I'll cancel the rest.

Now I have a special fondness for ammonia refrigeration systems:

• the first one was ammonia based
• Summer of 1968 I worked at an ice plant that used ammonia refrigeration
• One of the largest ammonia refrigeration explosions was in Oklahoma City where I grew up

Did I mention I was a mechanical engineering student in college? Thermodynamics including refrigeration and Carnot cycles, these are the things I find fascinating.

Bob Wilson

 

I have long wondered if that refrigeration cycle would work using exhaust heat?

Bob Wilson