# B Collision of stars

1. Sep 13, 2017 at 4:43 PM

### davidge

How frequently is collision of stars in galaxies? If there are 100 bi galaxies and 100 bi stars each, then how many collisions should we expect in within, say, 1 year?

2. Sep 13, 2017 at 6:05 PM

### davenn

that's a lower limit count

somewhere between zero and none at all

Don't you understand the HUGE distances between stars ?
The closest star to earth/our Sun is ~ 4.2 lightyears

They also don't move around the galaxy in a random/haphazard motion .... it is reasonably orderly

PS. you really should have labelled your thread with a B tag

Dave

3. Sep 13, 2017 at 6:07 PM

### Simon Peach

looks like a problem for the Drake equation, but different parameters!!!!

4. Sep 13, 2017 at 8:20 PM

### rootone

Gravity prefers orbits, collisions are possible but very unlikely

5. Sep 13, 2017 at 8:31 PM

### BenAS

Does a stellar merger count as a collision?

Edit, wiki has a page on stellar collisions
https://en.m.wikipedia.org/wiki/Stellar_collision

6. Sep 13, 2017 at 8:40 PM

### phinds

I'm pretty sure you have overstated the likelihood of it happening.

7. Sep 13, 2017 at 9:26 PM

### davenn

hahaha

8. Sep 13, 2017 at 10:57 PM

### stefan r

Someone should throw out numbers...

The sun's radius is 2.25461×10−8 parsec.
The local stellar density is about 0.14 stars per cubic parsec.
The sun move about 20km/s relative to neighborhood stars. one parsec in 1.5 x 1012 seconds which is about 49,000 years.
1.5 x 1012/ (2.25461×10−8 x 0.14)= 4.75 x 1020 seconds which is

1.5 x 1013 years for a collision.

With 100 billion stars in the milky way might happen in 150 years or so. very rough estimate. Leaning strongly toward none in a given year.

9. Sep 14, 2017 at 4:52 AM

### davenn

If you read the link by BenAS, you will find your estimates are out by a number of magnitudes

just a small clip ....
Now note that was for a globular cluster where star density is very high, particularly in towards the core. General galaxy stellar densities are substantially lower
other than maybe lose to the core and the massive black hole that is there.

Dave

10. Sep 14, 2017 at 9:56 AM

Staff Emeritus
And globular clusters do have evidence of collisions, or at least mass transfer (what does it mean for two gas balls to collide?). These are the so-called blue stragglers.

11. Sep 14, 2017 at 1:37 PM

### Chronos

There is a rare class of stars known as hypervelocity supernova, suspected to result from gravitational ejection of a binary star system by a black hole. Such an event can accelerate their orbital decay by orders of magnitude. Only a handful of candidates have been identified since their discovery in 2000. The death spiral of a binary system. which ordinarily requires billions of years, could be reduced to mere millions in the case of sling shot binaries. See https://doi.org/10.1093/mnras/stv789 for discussion.

Last edited: Sep 14, 2017 at 2:02 PM
12. Sep 14, 2017 at 7:25 PM

### stefan r

Did not square a number.

Is also not relevant. If the definition of "collision" is a merger than just coming within 2 solar radii is not a "collision". I thought my concussion was caused by a collision between my skull and pavement. Since my skull did not merge with the cement it must have been a close encounter.

N
~ 4.2 · D2 Myr−1
twice solar radius is 4.51 x 10-8 parsecs so
N ~ 8.54 x 10-21 encounters within 2 solar radii per year for our sun.

13. Sep 14, 2017 at 8:26 PM

### davenn

I really cannot make sense of that in the context of this thread

14. Sep 14, 2017 at 8:27 PM

### davenn

yup, indeed :)

15. Sep 15, 2017 at 9:47 AM

### TeethWhitener

Depending on your definition of "collision," you might only have to wait until 2022:
https://en.wikipedia.org/wiki/KIC_9832227
The two members of this contact binary in Cygnus will spiral inward to produce a luminous red nova sometime in early 2022 that will easily be visible with the naked eye (magnitude 2).