Radial attached and drag

[Jestersage]

Can someone point me to a topic that deals with the amount of drag incurred from radially attach to a parent, vs placing them on the node and then using the translate to move them around?

The above question is based on, assuming same parts, and same amount of exposed nodes.

[Foxster]

Generally, if you radially attach something then it has the full drag of the part. If you attach it in-line to a similar-sized part then the drag is reduced - and it makes no difference if you then translate it away with the offset tool - as long as don't rotate it. 

Edited November 16, 2019 by Foxster

[bewing]

For each part, the game uses an aerodynamic modeling to generate something called a "drag cube" and stores the info in the part database.  Depending on the orientation of the part, it has some drag coefficient calculated for it.

If the part is radially attached, then the drag cube causes drag, depending on the orientation.

If the part is node attached, then the drag cube of the part is reduced by the drag cube of the part that is next closest to the front of the craft, before calculating its drag (any offset is ignored).

 

Edited November 16, 2019 by bewing

[Jestersage]

10 hours ago, Foxster said:

Generally, if you radially attach something then it has the full drag of the part. If you attach it in-line to a similar-sized part then the drag is reduced - and it makes no difference if you then translate it away with the offset tool - as long as don't rotate it. 

So what if one DOES rotate it, say 180 degree?

[bewing]

The thing that makes the difference is the shape of the part compared to the airflow, and also compared to the part "in front" of it. If you take a 1.25m cylinder, and put it underneath another 1.25m cylinder and fly them through the air exactly prograde, then the lower cylinder is completely shielded from drag by the upper cylinder. If you rotate the lower cylinder 180 degrees, then it is still exactly the same shape, and will still have zero drag in the same circumstance. If you only rotate it 90 degrees, then the bottom cylinder is no longer the same shape as the upper cylinder, and will suffer some drag, but will be partially shielded from drag by the cylinder above it.

 

[Jestersage]

On 11/16/2019 at 7:08 PM, bewing said:

The thing that makes the difference is the shape of the part compared to the airflow, and also compared to the part "in front" of it. If you take a 1.25m cylinder, and put it underneath another 1.25m cylinder and fly them through the air exactly prograde, then the lower cylinder is completely shielded from drag by the upper cylinder. If you rotate the lower cylinder 180 degrees, then it is still exactly the same shape, and will still have zero drag in the same circumstance. If you only rotate it 90 degrees, then the bottom cylinder is no longer the same shape as the upper cylinder, and will suffer some drag, but will be partially shielded from drag by the cylinder above it.

 

Oh okay. So what about rotate 180 degree, and then translate (so basically "flip open")? Note that there are no exposed node points in this case.

Edited November 18, 2019 by Jestersage

[bewing]

27 minutes ago, Jestersage said:

Oh okay. So what about rotate 180 degree, and then translate (so basically "flip open")? Note that there are no exposed node points in this case.

For node attachment drag, translation has no effect. So a 180 degree rotate will have no drag.

[Jestersage]

Just now, bewing said:

For node attachment drag, translation has no effect. So a 180 degree rotate will have no drag.

Thanks. That means I do not need to rearrange my parts then.