Jump to content

[WIP] Nert's Dev Thread - Current: various updates


Nertea

Recommended Posts

I like the new NTR's and the new LH2 boiloff mechanism, but it looks like it's exposed an anomaly the NFE reactors..

When building an LH2-NTR-powered craft to head out an outer planet I noticed an anomaly with the NFE reactors producing power at warp.   As best as I can tell, you need battery capacity proportional to the power draw times the warp factor to buffer the power output of the reactor at warp.  

The upper stage was powered by the gas-core NTR (LV-N-GE 'Liberator') with fuel stored in a Mondo-60 and an HI-530.   Total power draw for cooling was about 130 Ec/s, well within the 400Ec/s capacity of an MX-4, so I added an MX-4 and a few radiators.   Had about 15km/s delta-V with good TWR.   Worked great until right after the burn for the interplanetary transfer, when I went to 100000x warp.   Stored Ec immediately dropped to zero, and boiloff started.

Aborted back to VAB, added about 400 Ec/s worth of solar panels (which wouldn't do in the outer system), and started experimenting in a 3000km circular orbit at varying warps.  With solar panels as the power source, batteries stayed topped up at warp (except when in eclipse).   With the MX-4 as the power source, it looks like stored power stabilizes at a deficit to full that was proportional to both the warp factor, and the power draw.     With only a few thousand Ec storage, going to higher warp (10000x and 100000x) blew through that in no time and sent the tanks into boiloff mode.

Looks like stored power stabilized with a deficit that was approximately (0.023 * warp-factor * consumption) below full.   So (for instance) with cooling enabled for both the Mondo-60 and the HI-530 it stabilized at about 298000 Ec below full.

So I tacked on a bit over 300000 Ec of batteries (!) and I finally had a ship that wouldn't boil off when the tanks were powered by the MX-4 at 100000x warp.

 

 

Link to comment
Share on other sites

14 minutes ago, billkerbinsky said:

I like the new NTR's and the new LH2 boiloff mechanism, but it looks like it's exposed an anomaly the NFE reactors..

When building an LH2-NTR-powered craft to head out an outer planet I noticed an anomaly with the NFE reactors producing power at warp.   As best as I can tell, you need battery capacity proportional to the power draw times the warp factor to buffer the power output of the reactor at warp.  

The upper stage was powered by the gas-core NTR (LV-N-GE 'Liberator') with fuel stored in a Mondo-60 and an HI-530.   Total power draw for cooling was about 130 Ec/s, well within the 400Ec/s capacity of an MX-4, so I added an MX-4 and a few radiators.   Had about 15km/s delta-V with good TWR.   Worked great until right after the burn for the interplanetary transfer, when I went to 100000x warp.   Stored Ec immediately dropped to zero, and boiloff started.

Aborted back to VAB, added about 400 Ec/s worth of solar panels (which wouldn't do in the outer system), and started experimenting in a 3000km circular orbit at varying warps.  With solar panels as the power source, batteries stayed topped up at warp (except when in eclipse).   With the MX-4 as the power source, it looks like stored power stabilizes at a deficit to full that was proportional to both the warp factor, and the power draw.     With only a few thousand Ec storage, going to higher warp (10000x and 100000x) blew through that in no time and sent the tanks into boiloff mode.

Looks like stored power stabilized with a deficit that was approximately (0.023 * warp-factor * consumption) below full.   So (for instance) with cooling enabled for both the Mondo-60 and the HI-530 it stabilized at about 298000 Ec below full.

So I tacked on a bit over 300000 Ec of batteries (!) and I finally had a ship that wouldn't boil off when the tanks were powered by the MX-4 at 100000x warp.

I dont think this is a weakness of the reactors as much as its a weakness of the stock analytic mode you enter at 1000x time warp and above. NF reactors have always had issues like this but since they were meant to power engines, its never been a problem until you introduce a sizeable power draw that sustains over warp such as boiloff refrigeration for substantially large craft.

Link to comment
Share on other sites

6 minutes ago, Captain Sierra said:

I dont think this is a weakness of the reactors as much as its a weakness of the stock analytic mode you enter at 1000x time warp and above. NF reactors have always had issues like this but since they were meant to power engines, its never been a problem until you introduce a sizeable power draw that sustains over warp such as boiloff refrigeration for substantially large craft.

The effect is observable (but not significant) even below 1000x timewarp.  It also doesn't affect solar power even at 100000x.   

Link to comment
Share on other sites

This has always been a problem, and I don't have much of an idea of how to fix it. With high time warp, it's a toss up whether the Ec production or consumption occurs first... let's say a solar panel adds 0.06 Ec/physics step, at 10000x that's only 600 Ec/tick. A medium reactor that adds 400 Ec/s adds 133,000 Ec per tick. In some situations that power is too large to get added to the ship's reserve so doesn't get added at all. This is actually why the reactors have built in Ec storage. The sequence goes something like this (or used to, dunno if recent KSP versions have adjusted this):

  1.  Reactor adds 400 Ec, room for only 300, none is added
  2. Tank consumes 50 Ec
  3. Reactor adds 400 Ec, room for 350, none is added
  4. Tank consumes 50 Ec
  5. Reactor adds 400 Ec, room for 400, power is added
  6. etc

So power will appear to get stuck at a level which the storage can handle the output. In the solar panel case, many small additions are more effective than one large one.

Someone can feel free to correct me about this, I'd like to fix it but I haven't been able to yet. 

Link to comment
Share on other sites

Can you see what the total energy consumption is somehow? If you could then just limit energy production to that plus a small amount to keep it from draining, instead of the full value of the reactor's max ec production.

Right now you force charging at the max reactor rate, but is that necessary as long as you charge just a tad faster than the load is taking away?

In case of a high load (ie. 1000ec/s or more) you will be focussed on the craft anyway and not in high warp. Just have the reactor produce 1000 + x ammount (ie 100) so batteries will still charge. When in warp there will only be a low load for sas/boiloff etc, say 20ec/s so the reactor will only supply load + 100ec/s. In both cases you will keep the batteties charged and also supply the present ec load up to the reactor's max rating. This smaller charge rate will prevent the issue you describe, but the reactor has enough capacity to ramp up and meet an increased load when needed.

Output = Load + x% with max output limited to reactors max value. Right now output = maxoutput all the time.

You can check battery level to see if it is depleting faster than you are charging and just ramp the reactor's actual ec output up or down

[edit] this is how it works in the real world too, your battery does not have zero impedance so you can't charge at unlimited (or super high) rate without increasing the voltage with the danger of charging so fast that you actually blow up the battery. All Ec sources in KSP seem to function like that however. For low Ec rated sources like solar panels or the NUKs this is not an issue to care about, but a reactor capable of delivering 2000-3000Ec/s it would be more realistic to charge the battery at a lower rate than the max Ec load possible, cq charging the batteries at 2000/s is not realistic. On an alternator the load gets it supply from the alternator (reactor in this case), not so much the battery.

 

Edited by Jimbodiah
Link to comment
Share on other sites

I'd like to make a request for a few more configurations of those excellent little RCS blocks from NF Spacecraft. Specifically, a version with nozzles pointing in only one direction (i.e. the block in the first screenshot without the side-pointing nozzles would be just what the doctor ordered). And, perhaps, if you feel like it, some more clusters like the one in the second shot. Yes, it's doable, but very fiddly with offsetting.

http://images.akamai.steamusercontent.com/ugc/357276631064647348/6244257604CA67FFB3B99FEF629DD0FDA1F78649/

http://images.akamai.steamusercontent.com/ugc/357276631064647877/73230C7DECA9839A8DA8C78CAFD68F530085DD28/

Link to comment
Share on other sites

Yeah these are probably planned. More RCS blocks at least. However, due to an abundance of RL stuff, my time for KSP is quite limited at the moment (my low post rate probably clues into this).

Anyways I'm also working on NTR support for NFE. The concept is as such: core temp controls ISP. You have to heat up the core before you start your burn. When the engine is on, the flow rate of your engine controls heat dissipated. You could add radiators to keep the core hot while not burning, or time the start of your burn so your reactor gets to Max heat when you are ready to burn.

The reactors will use a lot more uranium too, so keeping then on at full power all the time will be problematic. However, keeping them on at low power for the trimodal engines should easily be possible.

Link to comment
Share on other sites

Haven't been following development since December, or so, last year but Nert: but wow, just wow! That is some incredible modelling/texturing job you have done on those NTR's. Esp. since last time I've seen the previews of 2 models only, a joy to see.

I am not yet running ksp 1.0.5 (lazy to update), so still on 1.0.4 continuing my career save . So I expected some issues when testing this dev release on 1.0.4.

Two issues I have run into, don't know if they are related to my ksp version. Might be, so please do tell me :)

 

One regarding the models heat emission: It does not show.

The model only gets a dullish reddish glow, not like in the -e.dds texture. Has there been a change how heat emissions on models work between ksp 1.0.4 and 1.0.5? Or is this a know wip/issue still?

Seconds thing, and I know it is still in development, is boil-off plugin.

Also does seem to function on my end, no odd error s in the log regarding it. So is this plugin also non compatible with 1.0.4, or shouldn't it matter?

 

Thanks in advance.

Edited by Gkirmathal
Link to comment
Share on other sites

@Nertea, would you please recheck offline reactor's behavior in analytic mode? I have two more or less similar mining crafts, one with Garnet reactor, the other is on solar power. Even if the reactor is offline, I still find the craft overheating when returning to it. And this does not happen with the other craft. Same goes for offline reactor producing power on high warp.

Link to comment
Share on other sites

3 hours ago, Nertea said:

Yeah these are probably planned. More RCS blocks at least. However, due to an abundance of RL stuff, my time for KSP is quite limited at the moment (my low post rate probably clues into this).

Sure, sure, gotta take care of RL stuffs first.

Here's another configuration I think would be very useful for tanker/ATV type vehicles. Again, doable with offsetting, but very fiddly to do. Though I quite like the idea of a nozzle parts kit, and assembling you own blocks...

http://images.akamai.steamusercontent.com/ugc/357276631067213049/878419ECA953A0324116918E81F4B5885DF0F863/

Link to comment
Share on other sites

5 hours ago, Nertea said:

The heat and emissive animations use a number of parameters that were introduced in 1.05 so you might be out of luck there.

Figured it out, pre 1.0.5 works as a charm, though I might miss out on some bling. Al it needed was to edit the cfg to use ModuleAnimateHeat with * _HeatA animation.

 

The BoilOff is that restricted to 1.0.5 btw?

Link to comment
Share on other sites

19 hours ago, Psycho_zs said:

@Nertea, would you please recheck offline reactor's behavior in analytic mode? I have two more or less similar mining crafts, one with Garnet reactor, the other is on solar power. Even if the reactor is offline, I still find the craft overheating when returning to it. And this does not happen with the other craft. Same goes for offline reactor producing power on high warp.

You're going to need to provide some better information. I teleport a ship into orbit, turn on the reactor (or not), let it heat up, go back to tracking, warp a bit, go back to the ship, and nothing happens. Stick drills on it, nothing happens either. If I can't

Also FYI I can't touch anything in analytic mode. This idiotic heat catchup fix is a workaround for a stupid, unturnoffable stock system that I didn't even expect to work in the first place. 

Link to comment
Share on other sites

A ship with 10 drills, ISRU, 4 Jumbos, 4 Poodles, 1 Garnet and 9 XR-300 radiators or 8 Gigantors and 4 medium stock radiators.

I have a slight suspicion that putting drills and ISRU in actual use may have some part in provoking the problem.

Land on Mun or Minmus, deploy everything, start reactor, drill, fill up the tanks, retract everything, shutdown reactor, fly into orbit, forget about the ship for a day, return.

Edited by Psycho_zs
Link to comment
Share on other sites

Still unable to repro. Took a ship consisting of 1 garnet, 8 med thermal control systems, 2 drills, 1 isru, 1 jumbo-64 (emptied), a probe core and a couple of stack large batteries. Hyperedited it to a few places, filled the ore tank up with the drills, did a couple of toggles back to the space centre for some time warping, went back... converted some ore, turned the reactor on and off, some more toggles back and warping... could not create any anomalous heat. 

 

edit - just saw that post over in the release thread... you're not running FAR by any chance? Trying to figure out at least some correlations. 

Edited by Nertea
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...