Electric charge doesn't add up

[tomf]

I wanted to test what happened to the electic charge on with out of focus craft when they moved into shadow so I built a test probe with a single ox-stat, an octo2 core and a load of empty batteries. With the panel pointed away from the sun the drain rate was 0.03e/s as expected from the core.

With the panel pointed directly at the sun and right clicking on the panel showed an exposure of 1.0 and that it was generating .75 e/s. I would expect the charge rate to be 0.72e/s. However in resources the charge rate was only .54 e/s. Charging a 100 unit battery also took 185 seconds (.54e/s). Why does the panel report more charge than my craft seems to be receiving?

[Navy2k]

Are charging losses implemented in the Parts? I don't know but maybe a solution to your question. Because nothing is free in this world, not even charging your make believe battery in make believe space =)

Greetings

Ben

[RexKramer]

Any other electric loads? Was SAS on?

[tomf]

SAS was on but not providing any force. If it used power on a stationary craft I would have expected the requirement when facing away from kerbol to have been higher.

[Alshain]

Try turning it off, I think the SAS always provides some force, it's just minimal if it's isn't actively turning.

[tomf]

With SAS off I get the same thing

[Fr3Runn3r]

Could we get a picture of the craft to better understand what's going on? It could be that there's an electric charge sink that just didn't occur to you at the time.

[purpletarget]

Yes, a pic of the craft and readouts, and also it wouldn't hurt to see where it's located. The full output for solar runs on a roughly linear trend depending on the distance from Kerbol.

[bakanando]

Maybe it was charging the okto2 too? It has a small electricity storage that might have been also recharging.

[Alshain]

Maybe it was charging the okto2 too? It has a small electricity storage that might have been also recharging.

That's a good point. Should be easy to test too. Just stop the usage of the octo while it's full.

[tomf]

ok, here are two shots of a similar the craft, this time built with tthe cube core. In the first the panel is facing away from the sun to show the base load.

In the second then panel is getting the full .75 e/s

In the second the charge stored is only increasing by 0.63 when I expect 0.72.

[bakanando]

Can you click on the electric charge button on the resources panel so we can see the charge of the individual charge of batteries and core and take a pic just after activating batteries and one a couple seconds later? also, if it's possible try one core with only one battery instead of three and bonus points for putting another core with as many batteries as you can.

I'll test this tomorrow, as I have an exam today.

[The_Rocketeer]

Have you checked that the solar panel is definitely producing 0 e/s when facing away? If it's producing any charge at all it could be messing with your maths.

[bakanando]

Have you checked that the solar panel is definitely producing 0 e/s when facing away? If it's producing any charge at all it could be messing with your maths.

The core seems to be the only thing that should be using charge. The deficit shown on the first pic is right, the core uses 0.03E/s. The second is not, as it should only be using 0.03 for the core and everything else should be going to the batteries. Thus the surplus should be 0.72 and not 0.63 as is displayed in the second picture.

[EdFred]

Law of Conservation of Energy.

[bakanando]

I wanted to test what happened to the electic charge on with out of focus craft when they moved into shadow so I built a test probe with a single ox-stat, an octo2 core and a load of empty batteries. With the panel pointed away from the sun the drain rate was 0.03e/s as expected from the core.

With the panel pointed directly at the sun and right clicking on the panel showed an exposure of 1.0 and that it was generating .75 e/s. I would expect the charge rate to be 0.72e/s. However in resources the charge rate was only .54 e/s. Charging a 100 unit battery also took 185 seconds (.54e/s). Why does the panel report more charge than my craft seems to be receiving?

How many batteries were in your first probe? I'm guessing 6.

Law of Conservation of Energy.

Please be more specific.

Edited October 23, 2014 by bakanando
double bad grammar + reply

[The_Rocketeer]

Hmm... very odd.

It could be more than coincidence that you're effectively losing 0.03 e/s per charging battery (4 x 0.03 = 0.12; 0.75 - 0.12 = 0.63). You could test this by adding another pair of batteries and seeing if your total e gain is still 0.63 or if it changes to 0.57.

Apologies if this is dumbed down excessively, but I want to make sure you understand what I'm suggesting:

You would expect the game to deduct the draw from the incoming electricity (your 0.03 draw from the OKTO) BEFORE it distributes the charge to the batteries, but I guess what is actually happening is this draw is being deducted as part of the formula for each battery.

I.E.

It should be:

Charge rate = (Solar energy generated - Draw) / Number of batteries

But instead it's:

Charge rate = (Solar energy generated / Number of batteries) - Draw

If this is the case - well done, you've found a bug! Probably not hard to fix as it's just a simple formula error, but u should refer it to the relevant bug report channels.

Edited October 23, 2014 by The_Rocketeer

[The_Rocketeer]

How many batteries were in your first probe? I'm gessing 6.

I would have suspected 7.

(Soz double post)

[bakanando]

Hmm... very odd.

It could be more than coincidence that you're effectively losing 0.03 e/s per charging battery (4 x 0.03 = 0.12; 0.75 - 0.12 = 0.63). You could test this by adding another pair of batteries and seeing if your total e gain is still 0.63 or if it changes to 0.57.

Apologies if this is dumbed down excessively, but I want to make sure you understand what I'm suggesting:

You would expect the game to deduct the draw from the incoming electricity (your 0.03 draw from the OKTO) BEFORE it distributes the charge to the batteries, but I guess what is actually happening is a this draw is being deducted as part of the formula for each battery.

I.E.

It should be:

Charge rate = (Solar energy generated - Draw) / Number of batteries

But instead it's:

Charge rate = (Solar energy generated / Number of batteries) - Draw

If this is the case - well done, you've found a bug! Probably not hard to fix as it's just a simple formula error, but u should refer it to the relevant bug report channels.

That's what I was thinking too, but it'll need to be tested first.

I would have suspected 7.

Oops, I only counted three batts on the cube core, but there are four. (charge is 410) It seems strange though, it seems the four batts use 0.03 each but there's no core electricity reduction. Unless it's the CORE the one that is using 0.03 per battery.

And indeed, it might have been seven. I thought six+core.

Edited October 23, 2014 by bakanando
reply

[tomf]

Interesting idea that the bateries might be losing some charge but it doesn't look like it is the explanation, I just launched a test probe with 24 batteries and I am still getting 0.63

Has anyone else tried this, is it a bug in my slightly modded version/my experiment or the game?

First pointing away

Now pointing at the sun

Showing the charge status of lots of them

Which probe core I use might make a little difference

[The_Rocketeer]

I have just tested this with a similar craft. Strangely, I'm not getting the same issue:

I also tested with a radial solar array and got the same result:

I had a thought about the possibility of a symmetry glitch during build meaning you'd actually used 4 probe cores, but I guess you would see that as extra electrical storage on the ship too.

I suggest you try using other solar panels too, see if you still get the same weirdness with those.

Also, you mentioned that you were running a couple of mods - it might help if u can tell us what they are.

Frankly I'm otherwise at a loss.

Edited October 24, 2014 by The_Rocketeer

[Geschosskopf]

Why does the panel report more charge than my craft seems to be receiving?

KSP routinely violates conservation of mass and energy, so it follows that it should also violate conservation of charge.

[bakanando]

It'll help too if you can post which operative system (windows, mac, linux) are you using and whether it's 32 or 64 bit.

It's strange that you got the same reading with 4 and 24 batteries. Was the launch made with only one core or the rocket had more than one? Try going to space center and then coming back to your satellite to see if that fixes it. One last variable, was the core the root part of the ship?

Also, the one with 24 batts looks really nice.

[The_Rocketeer]

KSP routinely violates conservation of mass and energy, so it follows that it should also violate conservation of charge.

This would be true if I hadn't also tested the same scenario and had a different result. This doesn't appear to be 'routine'.

@bakanando: just for comparison, I'm running Windows 64-bit, stock install.

Edited October 24, 2014 by The_Rocketeer

[tomf]

For the last set of screenshots I unzipped a new installation and only installed hyperedit to make it easy to get the probe into space, so the probe never had any more parts attached and the core was the root part

I am running the windows 32 bit version.