# Are side boosters sometimes angled even if they are symmetrically arranged?

When more than one side booster is added to a first stage to increase thrust and they are distributed uniformly around the core (as opposed to this and this) thrust does not seem to need to be angled. They could all point straight down (except for dynamic thrust vectoring for steering) due to symmetry.

In the two images below, both from the recent NASA Spaceflight article Beidou-3 MEO-5 and MEO-6 launched by Long March 3B it looks to me like the exhaust from the boosters is angled outwards, I've added an annotated, cropped, and enlarged version to better show what I think that I'm seeing.

Is this in fact the case for this launch, and does this happen frequently?

[

• Thrust vectors are probably oriented to intersect the mean center of mass. That way, if one engine fails, you don't get a sudden pitch or yaw moment from the asymmetry. – Anthony X Jul 29 '18 at 4:50
• @AnthonyX I see, so approximately "torqueless". Losses due to $\cos(\theta)$ are small, and resulting radial loading due to $\sin(\theta)$ is manageable? – uhoh Jul 29 '18 at 4:54
• That configuration with the nozzles angled out was also used for the Ariane 1-4. – Hobbes Jul 29 '18 at 8:10
• @Jack Tangential gimbaling would be for steering control -- swing two opposing nozzles in the same direction and the vehicle pitches or yaws; swing them in opposite direction and it rolls. – Russell Borogove Jul 29 '18 at 17:08
• And the Snark – user20636 Jul 30 '18 at 9:21