Can extra-wide fairings actually reduce drag?

[herbal space program]

Hello all.

In the process of trying to design the lowest-dV Eve lifter possible (current personal best 5,240 m/s, 28t to LEO), I've come to appreciate how useful a well-designed fairing can be for reducing drag. One thing that kind of surprised me during this exercise is that when I made my 5m fairing wider than 5m by the smallest amount possible on its first segment, with the intent of protecting more aft parts from incineration on my fiery ascent, I found that it actually seemed to reduce overall drag! I expected to pay a price for that thermal protection, but instead it actually seemed to make my lifter accelerate faster than it did with a completely flush fairing. Is that an actual thing? By which I mean, can you actually reduce surface drag on more aft rocket parts by deflecting the flow outward with a slightly higher-diameter fairing above? It sure seemed like that to me, but I have no way of proving it. Thanks!

Edited February 26, 2019 by herbal space program

[OHara]

That is interesting.  The lego-style drag model was described, among other things, here (overhauls for 1.0).  

When neighbors are node-attached, each joining face its 'area' reduced by the 'area' of its mating part, for purposes of drag.  So a 1.3m² circular face of a fuel tank is completely shielded by the 1.4m² area of a reaction wheel, while the leftover 0.1m² area of the reaction wheel is left exposed.  The area of the entire silhouette of the part as seen from the staking direction counts as 'area'.

I haven't looked into it, but suspect the fairings are treated as if the procedural fairing was node-attached to the plate.  The 5-meter fairing has surface area 20.12m² on its largest face, 2.21m² on the others, according to my copy of PartDatabase.cfg, and this makes sense if it describes a 0.4-meter-thick base plate with 5.06-meter diameter (5.06m)² pi/4 = 20.12m².   

So maybe the slightly-over-5-meter fairing is enough to occlude the slightly-over-5-meter baseplate of that fairing.  The pointy fairing has a low coefficient of drag C_d, (the drag force is area×C_d×density×v²/2) and the flat plate has the largest possible C_d =1 so hiding every last bit of fairing plate with the procedural part would be a net win.

The alt-F12 menu lets you turn on "Display Aero data in Action Menus" which isn't very self-explanatory or documented, but might be useful to you.

Edited February 26, 2019 by OHara
typos matter more when they are digits

[FleshJeb]

4 hours ago, herbal space program said:

but I have no way of proving it.

Alt-F12 menu -> Physics -> Aero -> Show Aero Data in Action Menu

And the Aero Data GUI from the same menu.

This would probably help as well:

 

[herbal space program]

13 hours ago, OHara said:

The alt-F12 menu lets you turn on "Display Aero data in Action Menus" which isn't very self-explanatory or documented, but might be useful to you.

 

11 hours ago, FleshJeb said:

Alt-F12 menu -> Physics -> Aero -> Show Aero Data in Action Menu

And the Aero Data GUI from the same menu.

This would probably help as well:

 

 

Thanks! I will see what I can learn using those...

[FleshJeb]

53 minutes ago, herbal space program said:

Thanks! I will see what I can learn using those...

You're welcome. I'll be following this with interest since I have a similar use-case. The fairing was shockingly low drag (not shown), but I didn't look at the cockpit: