ReactorCraft - clever reactor setups?

[Ieldra]

About the batteries: the reason I used buffer batteries is that throughput calculations get very complicated if you put a lot of motors with different inputs on one battery. Maybe the code also doesn't things complicated . Also, yes, they're switchable by redstone. In fact, they only output with a redstone signal, though you can input at any time.

I asked Reika about where to pipe in the lubricant in the HP turbines. He said in the steam injector blocks, and that works. No idea where to go in the LP turbines since I never built any.

BTW, I added induction generators to the HP turbines and now I have 350ms ticks (14%) instead of 70% during the startup sequence of the reactor.

 

[Kirameki]

Awesome, thanks for the additional info. It could be they can't handle long buses, I didn't really consider it might *need* those intermediate buffers to behave. Worst case I can figure out a way to use a point to point and perhaps power a shaft bus instead with additional p2p to individual devices.
I run two LP turbines off of a HTGR as my current power source, plan to keep it around but idle once the fusion starts up as a backup system in case of catastrophe. (Note to self: Figure out if you can turn of the HTGR without removing all the fuel.)

 

[Ieldra]

I think I may need a little help with my fusion reactor.

First, it is running and it's producing about 26 GW power, and could probably generate more since my three turbines are maxed. But I still have problems with melting neutron absorbers and water does not appear to be the problem. The steam boilers have enough water but it appears the steam can't carry the heat away fast enough, since some of the boilers heat up to 750 degrees, and the neighboring neutron absorbers are alway the ones that melt. There also appears to a be a severely uneven distribution of neutrons. Some neutron absorbers stay at 250 degrees while others heat up to 1200 - and melt shortly thereafter. After these have melted - it's about 8-16 - things remain stable but this is unsatisfactory.

There is some added uncertainty because I see fusion plasma escaping the toroid, but I don't know if that's just the graphical glitch reported here and there of if there is something wrong with the reactor. Add to that, I have a tick rate of 13% (360ms). That tick lag started when I flipped the levers for powering the reactor up after adding the turbines and three induction generators to the system, and is almost unaffected by whether the reactor is generating power, the turbines running or not etc... Just adding three idle induction generators increased the tick lag while the plasma containment systems and the preheaters were running from 70% to 14%. Switching to vanilla graphics also had zero effect.

What can I do? Any ideas?

(It doesn't help that with tick lag like that, powering up the reactor takes about 90 minutes. Before I added the induction generators, that was about 15 minutes).

 

[Kirameki]

I think I may need a little help with my fusion reactor.

First, it is running and it's producing about 26 GW power, and could probably generate more since my three turbines are maxed. But I still have problems with melting neutron absorbers and water does not appear to be the problem. The steam boilers have enough water but it appears the steam can't carry the heat away fast enough, since some of the boilers heat up to 750 degrees, and the neighboring neutron absorbers are alway the ones that melt. There also appears to a be a severely uneven distribution of neutrons. Some neutron absorbers stay at 250 degrees while others heat up to 1200 - and melt shortly thereafter. After these have melted - it's about 8-16 - things remain stable but this is unsatisfactory.

There is some added uncertainty because I see fusion plasma escaping the toroid, but I don't know if that's just the graphical glitch reported here and there of if there is something wrong with the reactor. Add to that, I have a tick rate of 13% (360ms). That tick lag started when I flipped the levers for powering the reactor up after adding the turbines and three induction generators to the system, and is almost unaffected by whether the reactor is generating power, the turbines running or not etc... Just adding three idle induction generators increased the tick lag while the plasma containment systems and the preheaters were running from 70% to 14%. Switching to vanilla graphics also had zero effect.

What can I do? Any ideas?

(It doesn't help that with tick lag like that, powering up the reactor takes about 90 minutes. Before I added the induction generators, that was about 15 minutes).

What if you, as an experiment, add some steam grates near the offending boiler/absorber blocks to see if bleeding off excess steam will help resolve the issue? (Can just null the steam out into a condenser with your choice of fluid annihilation method.) If not you can at least potentially eliminate that as a cause. You may also want to try adding a second or third horizontal row of boilers on the outside to distribute the heat load over a wider area.
For the lag, have you tried checking an OPIS readout for possible specific culprits? I strongly suspect the generators, as you know from my other posts they've always caused problems for me in the past, unfortunately. Is the input to those generators fixed?

 

[Ieldra]

I think I have found the cause of the tick lag. I've posted things in my fusion reactor thread but I think this may be of general interest:

When an active (i.e. outputting power) Electricraft battery is connected to an input wire, its CPU load increases by a factor of somewhere between 10 and 100. It doesn't matter if energy is coming in on the input wire, just being connected causes lag. There appear to be aggravating conditions:
(1) The battery is outputting non-constant power.
(2) The battery is outputting constant power the amperage of which is not a power of 2.

Here's a comparison chart of two power systems for fusion reactor components. Note that disconnecting the HP turbine output line from the main auroral battery and routing the power to a second battery instead had the most drastic effect, even though no power was on that line when I tested it:

 

[Demosthenex]

I think I have found the cause of the tick lag. I've posted things in my fusion reactor thread but I think this may be of general interest:

When an active (i.e. outputting power) Electricraft battery is connected to an input wire, its CPU load increases by a factor of somewhere between 10 and 100. It doesn't matter if energy is coming in on the input wire, just being connected causes lag. There appear to be aggravating conditions:
(1) The battery is outputting non-constant power.
(2) The battery is outputting constant power the amperage of which is not a power of 2.

Here's a comparison chart of two power systems for fusion reactor components. Note that disconnecting the HP turbine output line from the main auroral battery and routing the power to a second battery instead had the most drastic effect, even though no power was on that line when I tested it:

Can you put a relay in front of the battery and the CPU be reduced? I'm just considering you have one charging, one discharging, and at an interval you flip via redstone signal.

 

[Ieldra]

Why the hell did you have to suggest that? Now I had to try it and for some weird reason the battery stopped outputting power and my pipe containment failed (that was why I originally had the buffer batteries). I am in the middle of a lava lake.

...and no, it does not make a difference. CPU load goes through the roof as soon as the input line is connected to a valid output device, regardless of relays and their state.

 

[Demosthenex]

Why the hell did you have to suggest that? Now I had to try it and for some weird reason the battery stopped outputting power and my pipe containment failed (that was why I originally had the buffer batteries). I am in the middle of a lava lake.

...and no, it does not make a difference. CPU load goes through the roof as soon as the input line is connected to a valid output device, regardless of relays and their state.

Was the CPU spike related to the explosion? I hear those can be laggy.

 

[Ieldra]

Was the CPU spike related to the explosion? I hear those can be laggy.

There was no explosion, just melting pipes. And no. I reloaded from a backup before the accident and tested this with a new power-up sequence.

 

[Kirameki]

I see I'm not the only one having problems with ElC. I've been experimenting with batteries and charging systems a bit, myself, I set up an induction from a LP turbine connected to 4 Auroral batteries via a length of 8 wires, initially adds about 10ms to lag, but overnight (8 hours) went up to adding 230ms to lag. Why the increasing levels, no clue. I haven't tried running the containment off the batteries for long-term yet, but ElC really doesn't seem usable in its current form unfortunately, despite being a great concept. I may just have to resort to using magnetostatics everywhere, which I was really hoping to avoid. Was considering using sticky pistons to make/break connections based on charge but the batteries don't seem to output to a comparator (at least in this version) so that won't work, either.

Correction: Apparently they DO output to a comparator, but -only- if the battery is active. How odd. Maybe you could try experimenting with this behavior?

 

[Pyure]

@LoaTBaC I see you removed your waste-eating reactor post. Was it ultimately a bust after all?

 

[LoaTBaC]

No, it still seems to work. Here it is: http://imgur.com/gallery/WFHk0/new. I thought I had necro'd the thread.

 

[Pyure]

Why'd you remove it?

I think its an excellent, positive example of @Reika's emergent-design philosophy. While it may need attention for balance reasons, at face value it seems to be a fun, interesting and plausible thing to do.

Could you add a screenshot of you confirming steam in the pipes?
Nevermind, I see that you did.

 

[LoaTBaC]

Why'd you remove it?

I think its an excellent, positive example of @Reika's emergent-design philosophy. While it may need attention for balance reasons, at face value it seems to be a fun, interesting and plausible thing to do.

Could you add a screenshot of you confirming steam in the pipes?
Nevermind, I see that you did.

Thanks. For those that didn't see it, my reactor design (http://imgur.com/gallery/WFHk0/new) uses neutrons that fly out of spent fuel containers to power a very small reactor core. The test one I build is able to sustain itself off a single fuel core, while still producing an acceptable amount of steam.

 

[Pyure]

@Demosthenex, @Ieldra may want to take a brief look.

btw: this is my thread, so I decide who necros it. I don't believe that "necroing" threads is a bad thing, so, I decide nobody necros it.

 

[Pyure]

btw I didn't know waste neutrons could emit vertically.

 

[LoaTBaC]

btw I didn't know waste neutrons could emit vertically.

Yeah, I was kind of surprised too. One nice thing about using neutrons like this is that, in theory, you could have some spent fuel containers beneath another fission reactor (or maybe even a breeder), and then use pistons to retract concrete shielding around them, bombarding the reactor with neutrons, allowing you to jumpstart the reactor, or temporarily increase the power level, almost like an afterburner for a jet engine.

 

[LoaTBaC]

Actually, you don't have to retract all the pistons at once. Since each time a neutron hits a block, it has a chance to stop, with virtually all neutrons stopping after four concrete blocks, you could possibly retract only some of the concrete, to throttle up or down the reactor, just like you do with Rotarycraft's gas and performance engines.

 

[Pyure]

That would work if ReC had a reactor temperature gauge like IC2 or BR or RealLife(tm), but I don't think it does. Temperature is actively "consumed" in chunks by cooling sources in iterative algorithms that makes it difficult to gauge and work with as is.

Reika wasn't interested in adding a second "true temperature" property we could use for this purpose. (I personally feel it would add immensely to the experience; he felt the benefit didn't justify the work, which is, obviously, his decision.)

 

[LoaTBaC]

That's a shame. I might try to do some testing with the idea, in case it might work better than expected.