DG cooling system

martins

Orbiter Founder
Orbiter Founder
Joined
Mar 31, 2008
Messages
2,448
Reaction score
462
Points
83
Website
orbit.medphys.ucl.ac.uk
I would like to implement a cooling subsystem for the delta-glider, partly to make the radiator do something useful, and partly to fill the rather blank right half of the overhead panel ...

To get me started, I'd like to somewhat model the cooling system on the space shuttle. I have a few questions regarding that:

  • Is the shuttle cooling circuit purely liquid, or does it include phase changes?
  • What are the typical temperatures of the coolant at the various positions of the loop, in particular on entering/exiting the radiators?
  • How much power is radiated off the radiators, and how much does the efficiency drop if the radiators are exposed to the sun?
  • What's the typical power consumption of the various subsystems, in particular avionics and life support?
  • Would it make sense to implement the cooling system with two loops, one for collecting heat from systems/cabin and dumping it in a heat reservoir/exchanger, and the other to pick up heat from the exchanger and pushing it through the radiators, or should it be a single loop?
  • Given that the DG radiators are rather undersized and can't always be deployed, would it be practical to also use the wings as radiators?
  • How much heat energy should be expected to propagate from the DG engines to the main fuselage during operation?
 

jedidia

shoemaker without legs
Addon Developer
Joined
Mar 19, 2008
Messages
10,891
Reaction score
2,141
Points
203
Location
between the planets
Is the shuttle cooling circuit purely liquid, or does it include phase changes?

Well, which one? The shuttle is a quite a merry amalgamation of temperature control systems... :shifty:
 

martins

Orbiter Founder
Orbiter Founder
Joined
Mar 31, 2008
Messages
2,448
Reaction score
462
Points
83
Website
orbit.medphys.ucl.ac.uk
Well, which one? The shuttle is a quite a merry amalgamation of temperature control systems... :shifty:

No idea. The .. big one? One you could envisage salvaging from a scrap Shuttle and retrofit in a DG? I rely on your best judgement. :lol:
 

boogabooga

Bug Crusher
Joined
Apr 16, 2011
Messages
2,999
Reaction score
1
Points
0
It's a good idea.

I would like to point out that the DG is an aircraft in addition to being a spacecraft. Unlike the real shuttle, it should be able to use its remaining fuel supply as a heat sink when the radiators are not deployed and especially during high-speed atmospheric flight.
 

boogabooga

Bug Crusher
Joined
Apr 16, 2011
Messages
2,999
Reaction score
1
Points
0
How much heat energy should be expected to propagate from the DG engines to the main fuselage during operation?

Impossible to say since the DG propulsion system is fantasy.
 

DaveS

Addon Developer
Addon Developer
Donator
Beta Tester
Joined
Feb 4, 2008
Messages
9,451
Reaction score
707
Points
203
It's a good idea.

I would like to point out that the DG is an aircraft in addition to being a spacecraft. Unlike the real shuttle, it should be able to use its remaining fuel supply as a heat sink when the radiators are not deployed and especially during high-speed atmospheric flight.
Actually, the shuttle uses a cold-soak of the radiators for cooling once the radiators are in bypass. This usually lasts until wheel-stop+10 minutes when the switch over to the ammonia boilers are required. This is part of the post-landing checklist, 5-8, RAD RECONFIG and NH3 ACT.
 

martins

Orbiter Founder
Orbiter Founder
Joined
Mar 31, 2008
Messages
2,448
Reaction score
462
Points
83
Website
orbit.medphys.ucl.ac.uk
I'd advise a read through of the SCOM: http://www.nasa.gov/centers/johnson/pdf/390651main_shuttle_crew_operations_manual.pdf, especially the ECLSS section. The Active Thermal Control System part begins on page 2.9-24.

Thanks for the link. I didn't see a reference to the coolant temperature in the radiators, but they do state that the maximum radiation power from the radiators is "61,100 Btu per hour" with translates to 1.79e4W, and the total usable surface area of the radiators is "1,195 square feet", which translates to 111m^2. Assuming black body radiation, that would point to a radiator temperature of 231K. That sounds surprisingly low, in particular since at another point they mention that the radiator outlet temperature is regulated to 38F (=276K) or 57F (=287K). Any suggestion as to the correct values? I just want to get an idea of ballpark numbers before starting on implementation.
 

Urwumpe

Not funny anymore
Addon Developer
Donator
Joined
Feb 6, 2008
Messages
37,656
Reaction score
2,378
Points
203
Location
Wolfsburg
Preferred Pronouns
Sire
The efficiency of the radiators isn't really high compared to a perfect black body.

Also, the Shuttle also uses evaporation cooling, either by flash evaporators (which use waste water) or by Ammonia boilers (which use a limited supply by ammonia, 49 lb per boiler). Flash evaporators only work above 100 Kft (about 30 km), below 30 km, the ammonia boilers are used.

The outlet temperature is regulated by mixing warm coolant with cold coolant from the radiators - the temperature of the radiators may be lower than the regulated output. I'll look for some design or simulation data, maybe this is more helpful.

About the power demand, you should of course include pump power, which is fairly easily calculated.

---------- Post added at 08:12 PM ---------- Previous post was at 08:09 PM ----------

This report might be already interesting about what other factors apply... also the inlet temperatures and predictions during a mission are shown.

http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/20070023916.pdf

---------- Post added at 08:14 PM ---------- Previous post was at 08:12 PM ----------

And here is a summary of the ATCS development and testing:

http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19740026227.pdf

Page 73 shows the simulation data that you are looking for, flow rates and temperatures at various measurement points.
 
Last edited:

jedidia

shoemaker without legs
Addon Developer
Joined
Mar 19, 2008
Messages
10,891
Reaction score
2,141
Points
203
Location
between the planets
Is the shuttle cooling circuit purely liquid, or does it include phase changes?

Most cooling systems that have to maintain room temperature have phase changes as far as I know, though I don't know exactly about the shuttle. I know that some of its cooling systems work by flash evaporation, which would definitely constitute a phase change... The real question is, how complex do you want to make the simulation. Simulating phase changes in the coolant might be a bit overkill.

Assuming black body radiation, that would point to a radiator temperature of 231K.

Doesn't sound that low. The thing is, for cooling the cabin you need quite a heat gradient from room temperature to coolant temperature. And of course from coolant temperature to the coolant that runs through the radiator, to get rid of the heat efficiently. I've done some work on cooling systems for IMS, although they are heavily simplified. The trouble is essentially that the hotter the radiator the better it will radiate, but the lower will be the temperature gradient you need to actually get the heat out of the cabin in the first place.
In other words, cabin cooling is the worst... Humans suck at getting hot.
That said, I don't even know whether the shuttle actually uses the main radiators to get rid of cabin heat, but I assume so. Cabin heat ends up requiring the largest radiator area because they have to run so damn cold.

How much power is radiated off the radiators, and how much does the efficiency drop if the radiators are exposed to the sun?

The efficiency would not just drop, it would effectively be anihilated (noticed that too when coding IMS...). Nasa gives the equilibrium temperature in earth orbit at 394 K, and as far as I can see they are calculating for a perfect blackbody here (no emissivity coefficient in the Stefan Blotzmann Law). In other words, any system you'd want to cool under these conditions has to be somewhat hotter if they are to transfer heat to the radiator. A bit too hot for the people in the cabin, I think...
Another problem here, though, is that you're not just getting heat from the sun... you're getting heat from earth, too, which is not to be underestimated, though I can't seem to find exact numbers at the moment.

Would it make sense to implement the cooling system with two loops, one for collecting heat from systems/cabin and dumping it in a heat reservoir/exchanger, and the other to pick up heat from the exchanger and pushing it through the radiators, or should it be a single loop?

In engineering, it's always two loops (at least). In code, it depends on how deep you want to go. I simplified to one loop, and people were still having a hard enough time understanding what's going on. Thermodynamics can be a bit counterintuitive when you grew up in a world where basically "we need to make it colder" means "throw more power at it". Your call how complex you want to make it.

Given that the DG radiators are rather undersized and can't always be deployed, would it be practical to also use the wings as radiators?

I'd expect them to be very much filled up with propellant... It's possible of course (at least on the upside... the bottom has to not turn to ash when reentering, which usually makes for very poor radiator material), but I'd expect the engineering compromises between tank fuselage, cooling fuselage, structural demands and wing profile to be a nightmare rarely seen before. Using remaining propellant as a temporary heatsink as boogabooga suggested would be a more sensible approach, depending on the nature of the fictional propellant.

How much heat energy should be expected to propagate from the DG engines to the main fuselage during operation?

This one's basically a compromise between ISP and heat... You can always reduce heat by reducing ISP a bit (pump more propellant through to carry away more heat). So what the question would come down to would be "how are the DGs engines optimised?" I'd suggest to do the math on the cooling system and then decide what it can take additionally, and just say that the engine's been optimised for that. Or just simplify it out. In IMS, I just let the engines run slightly hot over time, and if they're not of the self-cooling kind (electrical and fusion thrusters most prominently) they need their own cooling system and their own radiators anyways (radiators that run waaaay hotter than the ones needed to cool the rest of the systems).
 
Last edited:

martins

Orbiter Founder
Orbiter Founder
Joined
Mar 31, 2008
Messages
2,448
Reaction score
462
Points
83
Website
orbit.medphys.ucl.ac.uk
Doesn't sound that low. The thing is, for cooling the cabin you need quite a heat gradient from room temperature to coolant temperature. And of course from coolant temperature to the coolant that runs through the radiator, to get rid of the heat efficiently. I've done some work on cooling systems for IMS, although they are heavily simplified. The trouble is essentially that the hotter the radiator the better it will radiate, but the lower will be the temperature gradient you need to actually get the heat out of the cabin in the first place.

But if the coolant loop does involve a phase change, wouldn't you compress the coolant on entering the radiators to increase temperature and improve radiator efficiency, and expand it again on exit?
 

Thorsten

Active member
Joined
Dec 7, 2013
Messages
785
Reaction score
56
Points
43
Disclaimer: I have a fully working sim of the thermal physics of the Shuttle which solves the equations as described, so all is tried and tested.

Is the shuttle cooling circuit purely liquid, or does it include phase changes?

Liquid water loops for the hot avionics (which is air-fanned), dumping heat into liquid freon loops which also collect from cold planes where other equipment sits. Freon loops are either are cooled by the radiators, by the flash evaporators or by the ammonia boilers.

What are the typical temperatures of the coolant at the various positions of the loop, in particular on entering/exiting the radiators?

Freon/water exchanger tries to maintain 63 F, radiator outlet temperature is maintained to 38 F in low setting or 57 F in high setting by mixing freon flow through the radiators and by-passing the radiators. Which is to say, usually the radiator operates below its capabilities.

The rest of the temperatures is really a function of circumstances - how much power is on in the Shuttle, what's the current attitude to the sun and Earth,...

How much power is radiated off the radiators, and how much does the efficiency drop if the radiators are exposed to the sun?

You basically set up the radiative balance equations - heat load in the cabin raises freon temperature, the controller adjusts by-pass of the radiators to maintain the 38 F, that determines freon temperature in the radiator, incident sunlight and Earth IR radiation raises radiator temperature in addition, the radiator radiates at its blackbody temperature, that cools the freon,...

If you point the radiator into empty space and the Shuttle tail to sun, you get the freon overall a lot cooler than if you point the radiator sunward and have a large cross section to catch sunlight.



What's the typical power consumption of the various subsystems, in particular avionics and life support?

Shuttle runs on about 12-14 kW plus whatever payload needs. The 18 kW of radiator coolig capability thus are some extra for payload (and if you deploy the panels, I think you're getting another 20% on top of that).

I don't have a breakdown of all components, but

* cockpit lights are 1-2 kW at full illumination
* GPCs are ~500 W each
* heating elements of thrusters, hydraulics and fuel lines are often 100 W but not always on

I think the irreducible load with a bare minimum of systems must be some 7 kW.

Given that the DG radiators are rather undersized and can't always be deployed, would it be practical to also use the wings as radiators?

Radiator efficiency increases a lot if you make the loops run hotter...

If the wings are supposed to be radiators, they should not thermally be insulated - so then what do they do during entry? If they soak up the friction heat, you kill everyone...

How much heat energy should be expected to propagate from the DG engines to the main fuselage during operation?

Depends on what material is connecting them.

You have one heat reservoir of the engines, one of the rest of the glider. They are in thermal contact with a certain conductivity. As the engines heat, they will radiate heat off, and they will heat the rest of the ship via the conducting elements proportional to the temperature gradient and the conductivity- how fast both are depends on the masses being heated, the conductivity, the operating temperature of the engines,... If there's an insulator between them, the rest of the ship might not heat much at all.

You just have to set it up and simulate it, there's no simple answer.

Edit:

Here's a pic from a cold soak test in the sim (tail sunward to minimize solar heating) with a couple of temperature readings across the Shuttle fuselage and for various 'hot points':

Shuttle_coldsoak.jpg


Freon in temp is 310 K, freon out temp is 239, so the radiator does cool it quite a bit. I remember I checked that cooling the freon that low gives about the heat capacity to absorb for the right time when the radiator is not used during entry before the evaporators are needed (I was quite pleased with that...)
 
Last edited:

jedidia

shoemaker without legs
Addon Developer
Joined
Mar 19, 2008
Messages
10,891
Reaction score
2,141
Points
203
Location
between the planets
But if the coolant loop does involve a phase change, wouldn't you compress the coolant on entering the radiators to increase temperature and improve radiator efficiency, and expand it again on exit?

Yes, you would. As pointed out by Thorsten above, the temperature you have estimated is the exit temperature of the coolant, not the entry temperature. Was kind of confused on that myself yesterday, I wrote that post too late in the night after too much floorball.

Also, ignore this:

Nasa gives the equilibrium temperature in earth orbit at 394 K

I've looked at it in a wake state now, and realised that for some inexplicable reaseon, they completely forgott to take radiation into account. I thought yesterday that the value was way higher than what I remembered, but I thought "Hey, it's a Nasa source against your tired memory of simulations you ran almost half a decade ago, don't be stupid".
The actual blackbody equilibrium temperature at 1 AU seems to be somewhere at 280 K, which will still be too hot for something a lot less efficient than a blackbody supposed to cool something below room temperature. The real question is, how fast does the radiator heat up when exposed to direct sunlight, i.e. how fast will it become ineffective. I have no hard data on that, unfortunately. I remember just using the heat capacity of a sheet of aluminium in IMS to get something kind of credible (don't even remember how thick a sheet, probably around the 5 to 10mm mark).
 
Last edited:

Thorsten

Active member
Joined
Dec 7, 2013
Messages
785
Reaction score
56
Points
43
The real question is, how fast does the radiator heat up when exposed to direct sunlight, i.e. how fast will it become ineffective

It's a mirrored surface, which means it has a high optical albedo (also much of the fuselage is white, same thing) - it doesn't get heated much directly when exposed to the sun - IR radiation from Earth might be a different story. Just the Shuttle as a whole soaks up heat which eventually increases the load for the radiator.
 
Top