hey, super cool mod ! been enjoying it a lot :)

recently was playing with some nice setup of 4 reactors without ECCS. i use boilers for steam and more turbines with circuits to control temperature which is at 993 in middle reactor (973 on side), so at first i got 55 turbines but after it worked a while it turned out 55 turbines wasnt enough and steam was pilling up so now im at 56 turbines and power production is constant 504MW (goes down to 478 for just a second every minute but i think it doesnt matter)

it doesnt fit at all with what i read on forum and here ( 492 max from all 4 at temp 1000, right ? ) but i wouldnt report it only bc of it ... what really bothers me tho is that power read on reactors gui is constant 480 ( 2x118 sides and 2x122 mid ) so how i get that extra 24 MW ?

edit: i managed to settle it at 995 temperature with power production 504MW (every 90s it goes down to 460 for 1 second) but on reactors it shows 481 only

Yeah I'm getting actual power output higher than reported output as well, and it's messing things up a bit. My overheat protection keeps triggering, which shouldn't be needed.
In my case, I have 8 heat exchangers, all with full water and no steam backing up, that should be pulling 80MW of thermal power. Yet the reactor is reporting only 76 MW, and the whole system continues to get hotter! What gives? Should be cooling down at 4 MW in that configuration. It's like there's extra heat coming from somewhere.

Hi,

do you have a map where we can see that problem you describe?

Regards,

-max2344

Sure: https://drive.google.com/file/d/1r0cc28iVEXCadbjGWO02VaCpLBBpF_QJ/view?usp=sharing
I've added some labels and stuff to help you. You'll see I have a bunch of other mods as well, but for what it's worth I don't see how any of them could be affecting this.
This save is from a few seconds before the reactor reports 76 MW, which is the point that the pumps start feeding the 7th & 8th heat exchangers, taking the cooling by heat exchanger from 60 to 80 MW. This is intended to drop the power output back to 75MW, and keep around that level. Instead, when all 8 heat exchangers turn on, it keeps getting hotter.
Final note: when the steam storage hits 250k, it's going to shut itself down. (Just FYI - unrelated to the problem).

Hi Snoman314,

i guess you have to allow the download. I've sent a request.

Regards,

-max2344

Yeah I guess I set the permissions wrong. Try again now.

Hi Snoman314,

we've played around with your map for a while, however we could not find bugs in the reactor thermal production. We discussed your setup and the problems you are seeing in it for quite a while. The effects you are seeing are quite probably due to suboptimal design of heat/steam/ECCS network and quirky reactor control. Also we couldn't figure out a stabilization point in the system you designed. I've roughly fixed your setup. However we think you need to experiment more, therefore we won't tell you how to fix it :)

Regards,

-max2344

I can't figure out why it wasn't working. But I blueprinted it and made a copy, and that one works just fine, so I guess it's some weird bug in the base game? I literally built the exact same thing and now it works fine...

The problem came back after playing for a while, and saving/reloading a couple of times. So I got back to experimenting, and I still think the mod has a bug. This is what I thought through:

It's clearly nothing to do with the ECCS - that never even gets activated until long after the problem first occurs.

It's not an issue with the steam network - none of the heat exchangers ever have any steam stored in them, so the heat exchangers are never being limited by steam backlog.

It's not an issue with water supply to the heat exchangers - All the heat exchangers have 99% or so of their water buffers filled, and when all 8 exchangers are running, the water pump supplying everything is pumping 824 water per second, which is the exact consumption rate of 8x heat exchangers at full speed.

There does seem to be more power coming out of the reactor than is being reported - tracking the power output reported on the electrical grid I get more power than the reactor reports producing. I couldn't let this run long enough to fully stabilise due to lack of power demand, but it looked like it was going to settle on 80MW of electical power output, while the reactor was reporting 77 MW.

Finally, and most convincing to me - deconstructing and replacing the reactor fixes the problem. - Every other thing I changed made no difference, but doing nothing else but replacing the reactor itself immediately fixes the problem.

I've spent the better part of day working on this, so if it's something I'm doing then I give up. But It really seems like a bug.

Hi,

we looked through the code and played around with your map, but could not find any inconsistencies. The entities were behaving exactly like on our reference maps, but we used different technical solutions to build our reactor systems.

Building reactors in real life is not easy in general, so finding a working design might be difficult task as well. Also sub-optimal designs produce various problems as intended. YMMV.

Regards,

-dodo.the.last

If you're unable to pin down the bug, then fair enough. At least I have the workaround of deconstructing and re-constructing the reactor.

Hi,

we changed your design and the problem was gone, but we won't tell you how.

Regards,

-dodo.the.last

we changed your design and the problem was gone, but we won't tell you how.

Interesting. How do you explain the fact that the power level goes back to normal when I do nothing but remove the reactor core building, leaving all the heat pipes, exchangers, pumps, literally everything else untouched, then place the reactor back in the exact same spot, wire it up the exact same way, and start it back up?

You say you changed my design and fixed it. Well I can fix it without even changing the design (by re-building the reactor). Until I close the game and load it up again, then the problem comes back.

I have to say, it sounds like you're not really listening to what I'm saying.

Hi,

Does this behavior persist when you load your test map in vanilla and with Realistic Reactors only?

Regards,

-dodo.the.last

hello
i have same problem that after while perfectly balanced reactor goes to higher power for no reason

i build ton of accumulators before reactor and then go straight for 4 reactors at once (ingo formula) with system in place to enable/disable reactors when accu are at certain points
so i had some way to automate everything when reactors heat up to get them to desired temp, my design was working for weeks and then suddenly it all blow up ... after reloading save and trying to figure out what was going on it turned out reactors give even more power now so i had to rebuild and add few more turbines
it is my 2nd save with this cool mod and i remember on my 1st it happend to me too that after some time working it all blow up and i had to redesign, it was a bit diffrent story back then tho cos i had reactors very late game and they werent so much on/off like now when i have them early game
also so far (and it was same on 1st save too) after first change and redesign all is stable again, not like bug but more like some stat/formula changed in save

Hi trampek,

could you pls share a map where we can see what you mean? In general we haven't change anything in the code which could cause the behaviors described in this and similar threads.

Regards,

–max2344

sure ill share map but i dont know how reproduce it , it just changed power output and is how it is :P for now at least

sadly 1st save with situation i described in 1st post in this topic doesnt work now i cant load it so i think its no use to (i got blueprint of it if it helps)
i have 2nd save tho with my lowlvl reactor, after it exloded i was adding turbines until it stoped exploding :P didnt really care about design cos soon was going to upgrade it to tech3 turbines and stuff ... anyway it is this one that was working for weeks and suddenly go bum

https://1drv.ms/u/s!AsBgTBk0chmqgbFXIA98871wuuzNFg

Hi trampek,

thank you for sharing, we'll have a look at your setup. Not sure if a blueprint would help, depends from which century the designs are. Which RealisticReactors version was causing the change in the functioning of the reactor setup and which was it updated from?

Regards,

–max2344

hey
i managed to get into that 1st save but had to disable some mods (half base not working but reactor does) you can cleary see that power on reactor is lower than real energy output (like i described in 1st post here) also pretty sure update doesnt mess with it at all ... simple bc when i was playing this 1st save there was no update for year or so (other mods updated not this) and story was same - i build reactor it works alright for some time then bum , redesign and it works again but with more power.
https://1drv.ms/u/s!AsBgTBk0chmqgbFXIA98871wuuzNFg

just noticed some overflow valves are wrong direction , prolly cos mods updates so if you want it work for longer time you need correct em (arrow to right) they on right side of reactor 1 top and 1 down
also if you wonder why so much unused steam tanks ... those are only for that thing we discuss here if power go up for no reason once more they will be used and reactor disabled so i dont have load save ( which i had 1/hour cos long saving times)

Hi trampek,

have you tried reproducing that effect using vanilla/unmodded components?

Regards,

–max2344

no i didnt , as i told you it took lot time to happen and not feel like playing without mods so long just to make it happen ...

also it isnt any problem to me it working as it is, just reporting and trying to help but this bug or whatever isnt big deal to me

Hi trampek,

we have to say, your setup is quite delicate and a little bit too beefy for our hardware. We can't load it with all mods enabled, so we tried with RealisticReactors only on your map and were missing most of your reactor setup. So this leads us to believe to whole issue is related to the mods used in the setup. Could you provide a map version where only a minimum set of mods is used with your running reactor setup? Hopefully its small enough to load on our computers :P

Regards,

-dodo.the.last

thats bit wierd cos i dont have such good comp just some i5 with 16GB

anyway here it is https://1drv.ms/u/s!AsBgTBk0chmqgbFcINdc6JzcS7s_Xw

for some reason after i removed all other mods my accumulators are full when load so i place some around reactors so you can start it , 1 on right is hot cos i just stopped it so it will be faster to start but got no system in place to stop it when accumulators are full , you prolly need much more accumulators tho anyway if you want it run longer as without all mods nothing drains power anymore

when it start fully working you can easy see problem i described earlier - on reactors there is 119+122+122+119=482 but when you check power stats it turns out power from turbines is 512 on average

also i found 1 more thing - you can see that power output of this reactor changed when you check power stats , for long time output of turbines was 504 and then 512 (always 4 reactors, same target temperature) - that is what i was talking about - after sometime it started generate more power for no reason so i had to redesign it
i got no idea how reproduce it , same stuff happend with that 2nd save i gave you , there is no proof of it in power stats tho cos there i was getting power not only from nuclear but also from oil (same turbines so power stats tells nothing)

I confirm the problem. I analyzed it a bit and it generates some extra heat every 15 ticks. Things I’ve found so far:

1. The reactor consists of many entities of which relevant are two: realistic-reactor-normal and realistic-reactor-<power>.
2. Both have specific heat of 5MJ/°C.
3. Both are connected to heat pipes. (!)
4. The heat is generated by realistic-reactor-<power> by vanilla Factorio mechanic.
5. Temperature of realistic-reactor-normal is force-set by the mod.

I think here is the problem: realistic-reactor-normal was intended as purely display thing, but acts as a heat buffer, connected to the system, so force-changing its temperature changes overall energy content in the system. Oops.

A quickfix would be to reduce specific heat of realistic-reactor-normal to something tiny. But... are you sure you need that complicated setup? If you have to update every 15 ticks anyway, you could control energy production directly, increasing temperature by power*time/specific_heat instead of using a separate entity.

So, here is a version with two heat overproduction bugs hopefully fixed (tested on several setups):
https://mods.factorio.com/mod/UnrealisticReactors

numzero, would you be so kind to explain, how the two reactors connected with the heat pipes? I have myself made a couple of attempts of understanding the internal mechanics of the RR mod, but that particular aspect is still not solved for me. Just about any pointer will help.

Are reactors have some physical location on the map and the pipes are layed down between them or is it purely virtual?

Are reactors have some physical location on the map and the pipes are layed down between them or is it purely virtual?

snakeru, the two entities I’m talking about are both located on the map, at the exact same position. realistic-reactor-normal is the one you place, while realistic-reactor-<power> (where <power> is some integer) is the one RR adds for its internal needs. The prototypes are in prototypes/entities.lua, in reactor_normal and temp_reactor correspondingly, both based on reactor_template. Notably, both have the same heat_buffer.connections so, as the entities also have the same position, both connect to heat pipes at the exact same places. I don’t know how exactly does Factorio handle that situation but, they do transfer heat to each other (possibly through the surrounding heat pipes), which wasn’t intended.

Actually it’s not that hard to test. The problematic code for this specific case is in scripts/entity/reactor.lua, at line 885:

    -- UPDATE DISPLAYER --
    reactor.entity.temperature = reactor.core.temperature

If you replace it with e.g. reactor.entity.temperature = 950, you’ll see that in some builds the reactor will just keep everything around at near 950°C despite the comment stating it’s just for the display. It doesn’t affect all builds though, e.g. it depends on whether you place the reactor or the heat pipes first (which is especially bad as RR replaces the “core” (realistic-reactor-<power>) all the time).

I dont know if its relevant for you - or even correct: but here is some information from what I still remember from programing this years ago:
- The entity realistic-reactor-normal is burning fuel, and thus increasing its temperature, but not generating any heat (or almost any: reactor_normal.consumption = "0.00001W")
- The entities realistic-reactor-X are the invisible reactors which generate the heat. There is one entity for every generated MW, they will be placed by script according to the temperature of the realistic-reactor-normal. So for example: realistic-reactor-67 will be placed when realistic-reactor-normal has 140 degree (numbers are not correct, just as example...)
- This was necessary because (back then) you could not change the power output (i.e. heat generation) of reactors, this was the only way to get dynamically changing power outputs depending on some condition.
- From what I still understand from Factorio modding (and this is not much!) both realistic-reactor-normal and realistic-reactor-X entity have the heat_buffer property (inherited from reactor_template, line 253 in entities.lua). This could indeed lead to the problem that realistic-reactor-normal is absorbing and/or emitting heat, which was not intended.

Not sure how to fix this, but it looks like you guys have it under control ;)

Also keep in mind that if you place reactors directly next to each other, not only the heat buffer will be directly connected, but also the boilers from the ECCS, which might cause problems too during cooling.

Happy power generating...

Ingo,
First, thank you very much for the mod! It is much more interesting than vanilla reactors.

The entity realistic-reactor-normal is burning fuel, and thus increasing its temperature, but not generating any heat

heat=temperature*specific_heat, so “increasing temperature” and “generating heat” is one and the same. That said, with the power as low as specified, it indeed doesn’t generate any noticeable heat in this way.

you could not change the power output (i.e. heat generation) of reactors, this was the only way to get dynamically changing power outputs

To my knowledge the power output is still fixed in the prototype. However heat=temperature*specific_heat, so it’s possible to generate heat by just increasing temperature directly.

it looks like you guys have it under control ;)

Thanks. I hope so =)

if you place reactors directly next to each other, not only the heat buffer will be directly connected, but also the boilers from the ECCS, which might cause problems too during cooling.

Oops. Possible indeed. Will need to test.

So I published a version with these bugs actually fixed, not just worked around. No more entity dance, the heat is now produced directly. The core temperature fluctuates a bit (due to updates being every 15 ticks only) but, it works. You can see the source code on GitHub, as well as other info like my test setup.