Power units in game

[Paul Kingtiger]

I know this has been discussed before, but the last thread I can find on the subject is a year old and I don't want to be accused of necroing old threads so....

I've been looking into power generation a lot recently as part of my work on the Universal Storage mod. The mod uses real life units and performance as much as possible so it's important to get the power generation and usage right.

For parts within the mod that is easy, but it's also got to play nice with stock solar panels and other parts. I wanted to work out what one electrical unit in game is.

As a starting point I've used the ISS to gauge how efficient solar panels are in space. This site gives a area of 2500m2 and a power output of 85-120kw (I'll split the diffrence and say 100km). So that's 40watts per square meter (0.04kw)

The OX-4L 1x6 has an area of roughly 0.82m2 and produces 2 units per second.

Assuming the OX-4L uses ISS tech level cells, that's 32.8 watts.

So it would appear that 1 electrical unit = 16.4 watts.

I'm not really sure where the devs are coming from with this number. It's seems pretty much random, which makes balancing things a nightmare.

My suggestion is to change the electrical unit (next time you're releasing a save breaking update) to something closer to real world values.

For example, 1 unit could be 10 watts (or even better 1 watt) making calculations and balancing much easier.

[Red Iron Crown]

So 6 units is about 100W? Is the math really that much harder than 1 unit = 10W?

[Paul Kingtiger]

Actually it's worse!

I had a look at the other panels, and there is no consistency between the panels when looking at the number of panel units per m2.

Currently the large panels are less that half as efficient and the smallest ones. That just doesn't make sense.

presumably the larger panels should have the same solar cell technology as the small ones. I would understand if size made some difference in overall efficiency, either better or worse, but the difference is huge.

Edited April 11, 2014 by Paul Kingtiger

[frizzank]

Yea this is why I just spam the small cells all over my craft. The graph should be revised as the small ones do not break in atmo.

[karolus10]

This is true that using real units would be beneficial as You can use real life "stats" as base for balancing (you had something to compare), also they would be more readable as most people are accustomed to measurement units used in real life.

As far as I remember this was the first iteration of resources system and they won't changed as resources system has been cancelled for now... Units now are pretty much arbitrary.

[LethalDose]

So, what effect do you think this has on gameplay?

[Kasuha]

I don't think you can compare effectivity of real solar panels and Kerbal solar panels. If I'm not mistaken, even ISS has several generations of solar panels on it, each with different effectivity.

Also electricity production on KSP solar panels is measured at Kerbin distance from KSP Sun, while ISS solar panels are in rather larger distance from real Sun. We have also no data on TSI comparison for both stars.

But of course you can always use some reasonable guess. I would probably try to derive it from reaction wheels. We can measure their electricity consumption, their output is in kN/m and we can assume reasonable effectivity of reaction wheel drive when converting electricity to torque. I have no idea what will come out of it, though.

Edited April 11, 2014 by Kasuha

[Paul Kingtiger]

That assumes the reaction wheels are balanced. Although there is a ton of information online about real reaction wheels so it should be easy enough to check.

I do agree that you can't assume kerbal panels are the same as the ISS but it's a great place to start. Although the power per meter2 should be the same for every panel no matter the size.

[Kasuha]

The matter with solar panels is, their efficiency is pretty uncertain even on Earth.

[Paul Kingtiger]

If we placed the in-game panels on that chart the OX-STAT would be be right at the top and the Gigantor right at the bottom.

That's possible, although it doesn't explain why the large (and less efficient) panels are 2 tech levels above the small ones.

Also it's worth keeping in mind that all the panels in game are made by Probodobodyne, so would arguably use the same cell.

Ultimately I think the issue is that parts have been added to the game over a period of time without a balancing document to say that a part that does X, of size Y, will do it according to formula Z.

[Scoundrel]

1 EU in a battery is the equivalent of 1 coulomb - the reason it was left generic was to keep it simple as a resource for obvious reasons (well, obvious if you're an electrical engineer... Neomorph is the best person I know of to explain it). Unfortunately the thread I got that information from vanished in the Great Forum Derp of 2013, so there is no way to confirm it... but I did do some experiments involving rover wheels, batteries and very specific masses when they first came out and found that it's pretty darned close (and also that there's some surprising bugs/issues/quirks with the rover wheels that mostly go unnoticed, but that's neither here nor there). You should be able to do the same if you really need to know.

If I may digress for but a moment: the only real resource worth measuring is Snack Levels, as a closed loop life support system would function for a mere (estimated) 6-12 EU per kerbal per minute (assuming the less efficient Bosch reactor). A Sabatier reactor would require much less energy but would need constant hydrogen replenishment (read: sloooooowly sips liquid fuel) as it only utilizes half the hydrogen for water generation, requiring the rest to be dumped. If we assumed CMs contained enough liquid fuel dedicated to replenishing water and regenerative fuel cells to make enough power to provide that for life support as part of its design (which would help explain their noticeably disproportionate masses, but that's just a cognitive rationalization as to why parts are incredibly inconsistent), we would find that any non-Snack-based life support system have been made irrelevant.

That said, I really like your parts (I love the opening panels so we can see and poke at the guts with a stick!), and I get that there are people who like to micromanage everything they can, so for the sake of fun, you may have to abandon realism for consistency. Since 1 EU in a stored battery is 1 coulomb, you'll need to figure out how many volts the systems run on to calculate watts... which will likely reveal a noticeable disconnect between stock kerbal parts and the life support parts as the kerbalverse is a wee bit different from ours.