Cooling Convert-o-tron - I have this right?

[davidpsummers]

I have a question about cooling the ship pictured below (yes, there are some things wrong with it, but at the moment I just need to know about cooling).

The Convert-o-tron is running too hot.  But the three cooling panels are not being used 100%.  Is that because I'm using up all the Core Heat Xfer and the Cooling Percentage listed for the Max Cooling?  Ie. you can't get as much heat into it as you can cool (which seems dumb).  

 

[Streetwind]

3 hours ago, davidpsummers said:

The Convert-o-tron is running too hot.  But the three cooling panels are not being used 100%.

Counterintuitively, the "cooling %" readout on the radiators has nothing whatsoever to do with how much cooling capacity they are utilizing. What that readout really shows you is the radiator's temperature. In other words, this radiator is at 48.36% of its maximum temperature limit - probably around 1210K.

When looking at radiators, it's important to understand that KSP knows two forms of heat: "normal" heat, which governs the actual temperature of spacecraft parts, and "core" heat, which governs the pretend internal temperature of things like the Convert-O-Tron ISRU modules.

Normal heat is handled by the normal physical properties of the radiator part, including its actual size. It is pulled from other parts to the radiator once those other parts exceed (I believe) 400 K; the radiator raises its own temperature by lowering other parts' temperatures. Since it has a very high maximum temperature and great heat radiation abilities, it is much better suited to hold and slowly dissipate all that heat.

Core heat is handled exclusively by the "core transfer" stat on the radiator. It is not visibly processed by the radiator at all. It does not raise the radiator's temperature in any noticable way, nor is it affected by any of the radiator's own physical properties. It's a bit more complicated under the hood than the following, but for all practical intents and purposes, you can think of it like this: The game sums up all the "core transfer" values from all radiators you have installed. Then it checks if there is enough core transfer to handle the core heat produced by active ModuleCoreHeat partmodules. If there is enough, then the ModuleCoreHeat will hit their optimal temperature and stay there; if there is not enough, they will grow past their optimal temperature and lose efficiency.

In your case, you have three Thermal Control System (Small). Each has a 50 kW core transfer stat, for a total of 150 kW. However, the Convert-O-Tron 250 produces 200 kW core heat for each active process. So you do not have enough cooling to run even one, let alone two processes at the same time (as your screenshot shows). That the radiators' physical readouts do not suggest that they are at full load is completely irrelevant for core heat; those numbers only count for normal heat. All that counts for your use case is the sum of core transfer versus the ISRU's core heat production.

EDITed: Oh right, I forgot: the drills also produce core heat, and also need to be factored in.

Edited January 8, 2018 by Streetwind

[Cpt Kerbalkrunch]

@davidpsummers, you've got your answer above, which is the reason I prefer the medium thermal control. I usually add one per large drill (just a personal preference on the number; might be a bit much for you since you have 8 drills; I normally use 4). The only real advice I would add though, is to use an engineer. Looks like this is an unmanned miner that's meant to refuel itself, so it may not be feasible, but you'll see a tremendous difference in your drilling efficiency by adding a 4 or 5 star engineer. Drilling without them can be pretty tedious.

[bewing]

Actually, there seems to be a new bug involved here. I'll look into it further. A no-engineer conversion process should be running at about a 10% load, max (5% per channel). Which means 40kW core heat, max. Which two or three small TCS's shouldn't have any difficulty with.

 

 

Edited January 8, 2018 by bewing

[Geonovast]

2 minutes ago, bewing said:

Actually, there seems to be a new bug involved here. I'll look into it further. A no-engineer conversion process should be running at about a 10% load, max (5% per channel). Which means 40kW core heat, max. Which two or three small TCS's shouldn't have any difficulty with.

 

 

TCS uses EC, right?  Is it possible those RTGs aren't providing enough juice for the converter and the thermal stuff?

[bewing]

42 minutes ago, Geonovast said:

TCS uses EC, right?  Is it possible those RTGs aren't providing enough juice for the converter and the thermal stuff?

I thought that myself for a second, but no -- the converter uses EC too ... so if the craft ran out of juice, both of them would shut down at the same time.

What actually seems to be happening is that for this craft, the load percentage does not seem to be getting factored in. So even at 5% load (or .86% load), it's still producing 200kW of heat per channel.

 

Edited January 8, 2018 by bewing

[SpacedCowboy]

I'm certainly no expert on this but, could the problem be stemming from the fact that Lf+Ox AND Ox are being produced at the same time?

Edited January 8, 2018 by SpacedCowboy

[davidpsummers]

11 hours ago, Streetwind said:

 

EDITed: Oh right, I forgot: the drills also produce core heat, and also need to be factored in.

Thanks for the answer.  It was good.

Another question. The Ore is full.  The drills aren't bringing anything up.  Do they still produce heat or does one need to shut them off?

6 hours ago, Geonovast said:

TCS uses EC, right?  Is it possible those RTGs aren't providing enough juice for the converter and the thermal stuff?

When I looked at them, I didn't see that they produced any core heat?  (Ie. they have fins and cool themselves?)

Edited January 8, 2018 by davidpsummers

[Geonovast]

5 minutes ago, davidpsummers said:

When I looked at them, I didn't see that they produced any core heat?  (Ie. they have fins and cool themselves?)

The RTGs do cool themselves.  But the deployable Thermal Control Systems do use power.  I believe the canon is that they behave similar to liquid coolers for computers - gotta power the pump.

[bewing]

2 hours ago, davidpsummers said:

The Ore is full.  The drills aren't bringing anything up.  Do they still produce heat or does one need to shut them off?

No, they go into auto-standby mode and start cooling off.

 

[Kryxal]

If the drills are overheated, though, they WILL use the core cooling capacity.

[Neilski]

On 1/8/2018 at 2:44 PM, bewing said:

Actually, there seems to be a new bug involved here. I'll look into it further. A no-engineer conversion process should be running at about a 10% load, max (5% per channel). Which means 40kW core heat, max. Which two or three small TCS's shouldn't have any difficulty with.

 

 

I've just observed precisely this behaviour while testing conversion rates with 1/2/3 recipes running (i.e. a no-engineer 2-recipe conversion is causing overheating when it seems like it shouldn't).
Did you manage to look into the bug?

[bewing]

2 hours ago, Neilski said:

I've just observed precisely this behaviour while testing conversion rates with 1/2/3 recipes running (i.e. a no-engineer 2-recipe conversion is causing overheating when it seems like it shouldn't).
Did you manage to look into the bug?

Yeah, the devs changed the mechanics of it all. The ISRUs always produce the same heat now, no matter what the load is.

[vv3k70r]

On 1/8/2018 at 10:35 AM, Streetwind said:

the radiator raises its own temperature by lowering other parts' temperatures

And perfecting its heat exchange rate. Higher the contrast of temperature better HEX performances.

On 1/8/2018 at 10:34 PM, Geonovast said:

I believe the canon is that they behave similar to liquid coolers for computers - gotta power the pump.

Very reasonable if we manage heat in this simplified manner of active cooling throu fluid medium and pasive walls. It could be too complicated for gameplay to excercise buffers, flow, heat exchange between diferent fluid in radiators (one that take internal heat from parts, exchanging it to other medium inside leading to external radiator with all its buffers, fase transition and part stress) - it will be another piping on same complexity as for rocket engines or electric supply.

1 hour ago, bewing said:

Yeah, the devs changed the mechanics of it all. The ISRUs always produce the same heat now, no matter what the load is.

Is ambient temperature considered?

It would be cool to find out how many radiators are necessary on diferent celestial body.

 

Since we in this topic - how much cooling is necessary to get on 610km from Sun?

Edited January 1, 2021 by vv3k70r