Jump to content

Battery Transformer Infinite Power Loop


Recommended Posts

Is this fabled dark energy from dark matter? If we could drain power from dark matter to get dark energy in real life like that! Say good bye to global warming and energy crisis! We could make universe come back to us!

Link to comment
Share on other sites

On 2/21/2019 at 11:07 PM, mathmanican said:

How does it work? Not sure. (Edit: My current hypothesis is explained, with a formula, here.) I am guessing that the transformers try to pull from both the smart battery switching section, as well as the empty battery, and somehow successfully draw full power with half drawn from each. The build above effectively outputs 3kw from the transformers, but only drains 1.5kw (doubling the power in the system).  You can use the extra 1.5kw to power whatever you want.  Swap out the 6 transformers above for 5 big transformers (20kw of cycling power), and you can power 10kw of whatever you want (already tested it out). All of this can be done before cycle 10. Coal power?  Why?  Just use a wheel to insert a tiny bit of power into this thing, and then watch the power double rapidly (put the smart battery switch sensor really low to see it expontially fill).

Crazy?  Yep.  Anyone know if this has been seen before?  I know other bug reports are out there involving transformers pushing more power than they are supposed to push, but I don't recall them magically creating the power when trying to draw from an empty battery. Who needs infinte power from turbines anymore. Just use transformers and batteries. Haha.  Have fun tinkering.

Here's my theory, and it has to do with whatever algorithm they use for transformers..

Its been known for quite some time that you can run a grid at the capacity of a transformer and charge a battery at the same time.  Since the transformer only holds 1000J (4kJ for large T), this technically shouldn't be possible.  Its also possible to run slightly more devices than you should be able to provide power for (i.e. you can run 5 pumps on a transformer, though one will sometimes not run due to lack of power).

This is possible because there are two separate pieces to the algorithm for operating a transformer.  The first applies power to the consumers on the grid in question.  In the case of 5 pumps, your total draw is 1200 watts, or 1200J/s. The transformer only has 1000J "queued" up, so for the first tick only 4 pumps turn on.  This leaves a surplus of 40 Joules which is applied directly to the pump that is still waiting for power.  Assuming there is power available on the 'high' side of the transformer, the internal "queue" instantly refreshes to 1000J followed by a check of consumers.  As the 5th pump has some power, but not enough power, 220J is transferred to the 5th pump, giving it enough power to operate.  Any surplus power goes to batteries (or into the 'high' side of transfomers connected to the low side of the grid).  The last two ticks continue the same way: refresh transformer "queue" and check for power needs.  All four ticks complete in 1 second of game time.

The second check, as far as I can tell, is to calculate the average power draw of the consumers on the grid.  This only applies to consumers such as pumps or refrigerators.  Batteries are not part of this calculation.  5 pumps have an average power draw of 1200 watts, which is 200 watts more than the transformer can supply -- but this calculation appears to run over a larger time duration than the first part.  So for 1 second, 5 pumps are running drawing 1200 watts, which means that for the next second, one of the pumps MUST fail.  This drops the power draw to 960 watts.  This is below the maximum load that the transformer can supply power for, so at the start of the THIRD second, all 5 pumps can potentially operate.  The end result is that you get 4 pumps running continually and a 5th intermittently and burning out your wire.

The amount of power taken from the 'high' side of the grid is the sum of the power used by the consumers, plus the power taken up by any batteries (or other transformers) during the previous 4 ticks.

Now, here's where I think the extra power comes from, and its a combination of the above two properties combined with a third property of Oni electricity:  Power is simultaneously available to all parts of a given grid.  In your scenario that produces excess power, you have three transformers that are connected to three batteries and three more transformers.  So, 3x 1kJ of power is available to distribute to ANY component on the grid at the first tick of the first second.  Lets assume that you have sufficient power on the 'high' side of your three transformers.  During each tick, power is distributed first to consumers, and second to storage such as transformers or batteries.  The amount of power 'used' isn't calculated yet.  During the 4 ticks each second, your grid will allow 3 transformers * 1kJ * 4 ticks, or 12kJ of power into the grid.  3kJ of that power goes to the other transfomers, leaving 9kJ to charge whichever smart battery is on.  During the second check, it calculates how much power was used on the grid, in this case 9kJ, so 9kJ is removed from the 'high' side of the transformers -- which are connected to another set of transformers.  THESE transformers only remove 3kJ from their 'high' side because during the check of THEIR grid, only 3 transformers are needing power.  Thus 3kJ gets deducted from the 9kJ sent to the batteries, leaving 6kJ surplus.

In one second of operation, you've "produced" 6kJ of surplus power.

 

Side note: This only works if the batteries are never full.  If one fills up, that 'surplus' gets lost.

Link to comment
Share on other sites

2 hours ago, KittenIsAGeek said:

So, 3x 1kJ of power is available to distribute to ANY component on the grid at the first tick of the first second.

This line is the key problem.  Only 1 battery actually has power, yet power is being pulled from all 3. The real bug is this:

  • Transformers can pull from empty batteries, as if the battery is full, provided at least one battery on the grid has power. 

That's is. The bug is as simple as this. If they made sure that transformers could not pull from dead batteries, the problem would be solved. I've since built lots of different designs to draw power from dead batteries. For example, a manual shut off built in cycle 5 (once I have a small transformer), attached to a row of 30 tiny batteries (copper storage), can allow your dupes to fill one tiny battery, and then use the power from 30 batteries, provided you funnel all power through the transformer.  The transformer does not have to cycle back on itself.  Dead batteries provide power to transformers, provided one battery on the grid has juice. 

2 hours ago, KittenIsAGeek said:

Lets assume that you have sufficient power on the 'high' side of your three transformers.

Once this is true, you get NO extra power generation.  Power is treated normally. The system works as expected, and no bug exists. 

This bug should be quite simple to fix (whereas matter conversion for gases may not be). 

Link to comment
Share on other sites

Just now, mathmanican said:

This line is the key problem.  Only 1 battery actually has power, yet power is being pulled from all 3. The real bug is this:

  • Transformers can pull from empty batteries, as if the battery is full, provided at least one battery on the grid has power. 

That's is. The bug is as simple as this. If they made sure that transformers could not pull from dead batteries, the problem would be solved. I've since built lots of different designs to draw power from dead batteries. For example, a manual shut off built in cycle 5 (one I have a small transformer), attached to a row of 30 tiny batteries (copper storage), can allow your dupes to fill one tiny battery, and then use the power from 30 batteries, provided you funnel all power through the transformer.  The transformer does not have to cycle back on itself.  Dead batteries provide power to transformers, provided one battery on the grid has juice. 

Once this is true, you get NO extra power generation.  Power is treated normally. The system works as expected, and no bug exists. 

This bug should be quite simple to fix (whereas matter conversion for gases may not be). 

I think there's a bit more to it than that.  Transformers are a special case item because their "high" side is a sink, while their "low" side is a source.  If power is calculated by grid, this means that transformers get counted twice.  I believe that the exploit works because the empty battery is a proxy for the power of the transformer.

On the "low" side of the transformer, 1kJ of power is available at the beginning of each tick provided there is 1kJ of power apparently available on the grid.  This 1kJ of power can be transferred to anything on the "low" side of the grid, but can only be refilled if there is sufficient power to the high side of the grid.  A battery, on the other hand, can go either way: it can take the 1kJ from the transformer and make it appear available to the grid.  Both the transformer and the battery are required components of the exploit because the algorithm likely has a flag to set whether or not there is sufficient power on the 'high' side of the transformer while the calculations are going for the 'low' side of the grid.

The battery is dead and stays dead because after all the other calculations are done, there's no power left available to store.  However, the transformers stay full because during the calculations, that power was apparently available.

 

Link to comment
Share on other sites

8 hours ago, cpy said:

Is this fabled dark energy from dark matter? If we could drain power from dark matter to get dark energy in real life like that! Say good bye to global warming and energy crisis! We could make universe come back to us!

If we make the universe come back to us, we'll definitely have global warming.

...

And meteor showers. Meteors made of planets.

...

And time dilation from all the gravity.

...

And some other things.

Link to comment
Share on other sites

  • Developer

We took a look into this... the problem wasn't what we expected, and so it was really helpful to have this variety of scenarios and descriptions. Thanks!

(Note: it's still possible for batteries to charge eachother across a transformer, so in many of the setups here there will still be some oscillation while the system drains itself out.)

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

Please be aware that the content of this thread may be outdated and no longer applicable.

×
  • Create New...