Power distribution with battery on load side

[Sigma Cypher]

I have, or had, a bias against putting batteries on the load side of a circuit because ... overloads.

Recently I've been watching @JohnFrancis Base Lovin' videos again and I see some veteran players like Bob Terminator (does anyone know if he has an alias on this forum??) using batteries on the load side this this:

I am just guessing about the automation.  Bob didn't have the power shutoff on the right.  I assume he does this to minimize heat from the transformer when it is not needed, because I don't see another reason.  

Another regular poster on this forum said she uses batteries on the load side too, but hooks up the automation like this:

because (she said) doing so preserves the energy in the transformer, while shutting it off directly makes it go to zero.  This was a few updates ago.  Now I think the behavior is different because I observed that it is the other way around:

Which kind of makes sense, because the energy remaining in the transformer gets drained into the battery almost instantly when the shutoff is opened (but transformer still on), whereas if the transformer is turned off it can't deliver its energy.  It does however run-off eventually --  I would guess at about 4J/tick with the large and 1J/tick with the small power transformer contrary to what the wiki says).  If there is no load on the battery, the two automation configurations are identical: both result in the eventual loss of the energy in the transformer due to run-off.

So ... can anyone give me a reason to start putting batteries on the load side for any reason other than say trying to run something like this with a small power transformer:

: 

(not even sure this works honestly).

Thanks!

[psusi]

Did they explain why in their videos?  I don't see any reason for it.

[nakomaru]

47 minutes ago, Sigma Cypher said:

because (she said) doing so preserves the energy in the transformer, while shutting it off directly makes it go to zero.  This was a few updates ago. Now I think the behavior is different

Correct. This bug was fixed. Now you can disable a transformer by disconnecting the the power anywhere, or disabling it by automation, and it will not delete its charge in any circumstance.

47 minutes ago, Sigma Cypher said:

So ... can anyone give me a reason to start putting batteries on the load side for any reason

As presented, I don't see much reason. I would guess the reason they were doing this is because they have a switching battery grid. This lets you put any amount of wattage through any power line by using transformers to charge batteries, then disconnect from the grid (purple) and connect to the consumers (white) when charged.

Grid supply (purple) becomes something like this:

Spoiler

[Nedix]

I use a battery on the load side for several reasons:

1. It happens that I push the super conductive wire over the 2k limit and a battery helps for avoiding overload for a limited time. This is not a usual norm but it happens at least until I optimize the circuits

2. Heat generation - a large power transformer has a 1k DTU/s heat generation while a smart battery only 500. In plenty of scenarios this makes quite a difference

3. Connected to #2 - I have plenty of devices which to not consume power 24/7 so it really makes no sense to power them all the time. I'd rather live with the power leak than with setting up a cooling solution for 24/7 transformers

 

I don't use the battery on the load side only in the late game when I have plenty of man-power, resources, power generation etc. that I can afford big power boxes with dedicated cooling.

[TheMule]

One setup I like to build as soon as I have plastic and steel (it can be done with golden amalgam) and crude oil, is a CSV tamer with a metal refinery.
Usually this happens if I find a CSV right after early game. I don't really care for 100% uptime, even tho with not too complicated bypass pumps you can achieve that early on (you do need 2 liquids that don't evaporate at 200C, usually crude oil and petroleum), I want some water source to plug my early SPOM to.

Sometimes I use heavy watt wire (need vacuum not to leak heat thru the plate and can be annoying) but sometimes I use 2 1kW transformers with batteries on the consumer side, one for the refinery, one for the AT. And I connect everything with conductive wire (2kW). It's usually self powering if I refine steel or iron mostly. Neither the refinery nor the AT run continuously. The batteries on their side are enough to run them and while they're inactive the turbine recharges the batteries (slowly). But again the heavy watt wire setup isn't that complicated. It more the decor thing that bothers me but you can use gold heavy watt conductive wire next to the refinery to counter that. You still need batteries of course.

[Sigma Cypher]

@psusi No, John didn't go into that.  Bob was consistent about it though.  Every single transformer had a battery right next to it on the load side.

@nakomaru Why not use regular (1kW) wire in your purple supply grid?  The load is zero I think?  Up until the power shutoffs anyway.

Thanks @Nedix.  I see that it makes sense to skip the transformer all together in some scenarios.  I guess I sort of forgot that and just plop down a transformer between generators and load as soon as I unlock the tech without thinking.  I don't understand #1.  This seems odd, maybe even backwards to me, but I confess that I don't understand why I sometimes don't get damage when the circuit is intermittently overloaded.  To avoid pushing 2k wire over the limit I have used two small transformers in parallel.

EDIT: Isn't it time Klei "fixed" the large power transformer to limit the load to 2kW or gave us a 4kW wire?  I have been away for a while but I figured for sure this would be fixed by release date.

[nakomaru]

52 minutes ago, Sigma Cypher said:

Why not use regular (1kW) wire in your purple supply grid?  The load is zero I think?  Up until the power shutoffs anyway.

You are free to. I use conductive wire because it looks better, has better decor, and allows me to overload the main grid by 2kW when constructing new 2kW consumer stations before I worry about finishing the battery switcher prior.

[Nedix]

@Sigma Cypher If I remember correctly since there is no energy buffer between the transformer and the consumers then when you exceed the load on the wire you immediately start damaging it. If you have a battery in between then it can sustain the overload for a while. I think this dates back to the early days of ONI and I just made it a habit.

PS: I also use a smart battery on the load side when I have coal generators and control them through the automation wire so they burn coal as much as they need to load up the batteries and then they stop.

[nakomaru]

1 hour ago, Nedix said:

If I remember correctly since there is no energy buffer between the transformer and the consumers then when you exceed the load on the wire you immediately start damaging it. If you have a battery in between then it can sustain the overload for a while

This is kind of backwards. If you have a battery connected to consumers, it will be able to supply any wattage. Damage is only determined by consumed wattage, which means a battery will be able to supply >2kW and damage it. A transformer is limited to e.g. 2x1kW however.

(Previously, transformers COULD briefly supply 5x their rated power, because it could release e.g. 1kJ per tick = 5kW while it was charged. It was only properly limited on its charge side. Now it's properly limited on its output as well.)

But you don't really need to do that if you design 2kW consumer grids.

[psusi]

1 hour ago, Nedix said:

1. It happens that I push the super conductive wire over the 2k limit and a battery helps for avoiding overload for a limited time. This is not a usual norm but it happens at least until I optimize the circuits

No it doesn't.  If you have more than 2k of load on the wire, it is overloaded and will take damage.

 

[Nedix]

@nakomaru, @psusi I probably misunderstood the whole thing but in the end it's still a habit