Major Mysteriously Unpowered Cables

[Studenterhue]

A very strange, very inexplicable bug, where the station's entire network of cables appear to be unpowered when pulsed by a multitool. Eventually, as APC cells lose charge, and with the wires apparently unpowered, various rooms in the station lose power about 30-40 minutes in, one by one.

You can alleviate the power loss somewhat by switching the APC cells when they lose charge. The cell chargers will still be running, so you can counter the electricity shatting itself by casually violating the Second Law of Thermodynamics. Hence, why I put Major rather than Critical. 

I really do not know how to replicate this. Some rounds it happens, some rounds it doesn't. All I have are some observations:

• I really do mean they're unpowered. Even the wires directly below the TEG or solar panels will appear unpowered. 
  
• Occurs regardless of whether or not the engine (the TEG in most of my experiences of this bug) is hotwired, or even running. Mild charburn or face-melting hellburn, the multitool will still say they're unpowered.
  
• Seems to occurs regardless of whether the solars are hotwired, regardless of whether the solar SMES Units are configured. (Memory of this is bit spotty, because I usually hotwire first, check the power status, then check the SMES).
  
• Heck, it (definitely) occurs even if the SMES Units are set to output power. Though memory is spotty over whether the SMES Units said they were actually outputting power. 
  
• The SMES Units will still be charging though, somehow.
  
• Pretty much all of the station APCs will say they are not receiving external power.
  
• However, most of the machines will still be running, the lights will still be on, the emergency lights will turn on if Red Alert is declared, etc.
  
• Cutting cables you'd expect to be powered still releases sparks, even though the multitool will say they're unpowered.
  

[Cyfarfod]

Oh yeah, I can add some info, I pulled this off a few rounds ago. He's what I did differently from a normal hellburn, it's pretty quantifiable actually.

One, I went to the bottom-most solar array SMES, in cog1, and turned up it's charging and output.

Two, I let my combustion chamber heat for a full 26 minutes before starting the hot loop. I go to 20 minutes as a rule but I'm experimenting.

Three, I put a portable air pump on the aux port (hot loop) below the TEG, bottom left of that room. Had it at max KpA, set to in, and turned on.

Two final notes- APCs were losing power but not, IME, as fast as they SHOULD have. And, while for the vast majority of the round the ONLY way to see the TEG output was the indicator on the engine itself (power computers showed 0kw output and no pressure/temp information, same for Power Checker on PDA) towards the end of the round Power Checker was able to display info finally. This was, I believe, not until shuttle had arrived,

[New525]

Ive seen this happen more than once

[Cyfarfod]

My gut is it's one of two things-

One, superheating too small a volume of gas SERIOUSLY fudges up things. We know this can be true for another system- trying to fill tanks from canisters of incredibly hot gas. It just won't flow out because... I dunno, I just picture the atoms are too big or something. In the example I said above, my combustion chamber was at... I think about a billion degrees before I even started gas in. This meant it expanded like crazy, and with the pump pulling it back OUT of the system after being cooled by contact with the cold loop that probably didn't help either.

Second, this is anecdotal, but I feel like there have been plenty of times on various maps where I latejoin as CE (one of my fav jobs, if I see no engineers and CE is open I'll almost always take it) and there seems to be a certain point where those wires aren't gonna take power no matter what you do.

EDIT: Next time I'm playing during the lowpop hours I usually see right before MY bedtime, I'll try a few things and see what I can pin down, maybe I'll get something useful.

[Cyfarfod]

Alright, last night/this morning I was able to make attempt one to repro. I once again set up the engine my usual way, same as in the example before where the cables broke, with the air pump removing gas from hot loop AFTER it leaves TEG. Let combustion chamber run to 26 minutes, DID NOT do anything with Solars this time. Nothing unusual. I'll try a few more times without touching solars, then do it with, and edit the results into this post. Have a hunch it's an interaction between generator and solars that are causing the problem though.

EDIT: Tried another burn the following night, as identical as I could realistically do to the one above but this time had an AI set up ONE solar to produce extra power. Cables did not lose power. Next time I'll do the solar myself, that was poor rigor to be honest, having someone else do it.

[Dr Zoidcrab]

Which map(s) is this occuring on? Might be worth noting that some of the solars start hotwired on certain maps. (arrivals/escape on cog2).

Might be an interaction between having a SMES running and and a hotwire at the same time.

Try hotwiring multiple power sources/bypassing the SMES, i've had the powernet lock up when I tried to hotwire each SMES room on donut. See if disabling the smes or removing any hotwired solars fixes it.

[Superlagg]

This seems to still be a thing, at least on Cog1. Happened twice that I noticed. First was when I got the engine good and toasty and turned on the PTL. It ran alright for a while, but around 500GW it just stopped working. Checked the wires from the engine on out past the smesses, and they were unpowered. Same thing a second time later as an antag CE, got the engine good and murdery, but the cables got unpowered pretty early on, around 1GW.

[Triacontakai]

Coders are aware of this afaik, theres actually a verb specifically to restart and fix the powernet system since its funky as hell.

[Recusor]

If this happens is it appropriate to ahelp to request that verb be used?

[Vengarr]

If this happens is it appropriate to ahelp to request that verb be used?

This happened today on Oshan and an admin reset it when I ahelped.