Electric charge doesn't add up

[The_Rocketeer]

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.

That's a difference. I didn't use hyperedit.

I wonder if something happens in the game's back end when you 'launch' the flight using the spacebar that would change the charge? All total guesswork, but perhaps you should try launching an actual rocket and see if it still exhibits the same problem.

[bakanando]

I think I got the same error this time, but with a spaceplane I had already on orbit. After some use of the reaction wheels at night I had used some power and the moment the sun hit the panels (three/four with a rate of around 1.4 total) the rate of charge was displayed as 0.58 with and without SAS enabled at the time.

Dunno how to post the pic, so I'll just put the link here:

http://steamcommunity.com/sharedfiles/filedetails/?id=331379837

oh, I forgot. Linux 32-bit. Running through steam, vanilla install.

[The_Rocketeer]

Weird. I can't get my head around this one. I'd be tempted to make a bug report anyway because this doesn't feel right at all.

Incidentally, you can post the image by using the Insert Image tool on the reply window (third on the right from the Link tool). Then just paste in the URL.

[Padishar]

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

It only goes "wrong" when you are charging a part that is already full of charge (and is enabled). The more parts you have that are full and enabled, the less charge goes into the non-full parts. This appears to be a bug (or a very strange intended feature) of the Part.RequestResource function (this is used to both remove and add a resource to a part) for resources that use either ALL_VESSEL or STAGE_PRIORITY_FLOW.

Incidentally, you can do exactly the same tests on the launchpad using an RTG instead of a solar panel (you may need to tweak the .cfg of the RTG to make it not "always on" and then set up an action group to toggle it).

If you have a ship with an empty core and an empty z-100 battery and disable either of them then you get the correct charge rate of 0.72 (0.75 - 0.03). If you leave both of them enabled then the charge rate is initially correct at 0.72, but as soon as the probe core becomes full the charge rate drops to 0.63.

Edited October 29, 2014 by Padishar

[cantab]

So what this sounds like is that a full, enabled battery still takes in charge but then just dumps it.

Which is actually somewhat realistic. A full battery connected to a charging power supply will "dump" that overcharging as heat into the battery.

[The_Rocketeer]

It only goes "wrong" when you are charging a part that is already full of charge (and is enabled). The more parts you have that are full and enabled, the less charge goes into the non-full parts. This appears to be a bug (or a very strange intended feature) of the Part.RequestResource function (this is used to both remove and add a resource to a part) for resources that use either ALL_VESSEL or STAGE_PRIORITY_FLOW.

Incidentally, you can do exactly the same tests on the launchpad using an RTG instead of a solar panel (you may need to tweak the .cfg of the RTG to make it not "always on" and then set up an action group to toggle it).

If you have a ship with an empty core and an empty z-100 battery and disable either of them then you get the correct charge rate of 0.72 (0.75 - 0.03). If you leave both of them enabled then the charge rate is initially correct at 0.72, but as soon as the probe core becomes full the charge rate drops to 0.63.

A very good theory... but.

These are discharged batteries in tomf's demonstration craft (see resource display top right). Based on your info, they should still be charging at the full rate.

EDIT: Just re-read your last paragraph. That could certianly be the issue. Tomf?

Edited October 29, 2014 by The_Rocketeer

[Frank_G]

Hm... the only thing i can think of is, that you denied energy usage... if you switch off batteries, you cant recharge them either.

[Padishar]

A very good theory... but.

http://i.imgur.com/T9AT95L.png

These are discharged batteries in tomf's demonstration craft (see resource display top right). Based on your info, they should still be charging at the full rate.

EDIT: Just re-read your last paragraph. That could certianly be the issue. Tomf?

Yes, I believe in that picture, the probe core is full and enabled so one "chunk" of charge is going missing. If you disable the full probe core then the charge rate will jump up to 0.72...

However, it isn't a simple matter of the charge being shared equally between the enabled parts or a lot more would be going missing. It seems to have something odd to do with the fact that the probe core (and all the enabled batteries) is slightly discharged on every frame (by the probe core usage) and also recharged by a part of its share of the charging but even this doesn't correctly explain the numbers...

Edited October 29, 2014 by Padishar

[tomf]

I checked this by using a probe that had 12 full batteries, a core and 1 empty battery and the charge rate was down to 0.14 so it looks like Padishar is correct.

I can't work out the maths though, it certainly isn't the available charge being divided equaly between charge stores.

[bakanando]

In my case, both batteries and core were not full and it still had only -0.58 of charge. But it does look like the issue is somewhere there.