Let's go hump hunting!

The place to discuss particle physics
Kasuha
Posts: 570
Joined: Tue Dec 15, 2009 1:22 pm

Re: Let's go hump hunting!

Post by Kasuha » Tue Apr 06, 2010 1:38 pm

Tau wrote:(otherwise it would influence both beams, for example, not just one)
Oh. This is new to me, I always thought both beams are affected.
Well, this really pretty much rules out global or distant sources.

User avatar
Tau
LHCPortal Guru
Posts: 168
Joined: Tue Jan 05, 2010 10:16 pm
Location: Heemskerk, Netherlands

More news

Post by Tau » Tue Apr 06, 2010 7:58 pm

The hump scan they did today gives a lot of new info for wannabee hump hunters. I've been thinking about them, but I can't really figure it out.
There's something going on with sum and difference frequencies between the tune and the hump here. Some (but not all) peaks add up to exactly .5; I guess this is an artefact of the FFT acquisition, but I am not sure.
But there's so much info in these graphs, this should help in figuring out what the hump does.
I noted also that the hump does almost exactly the same thing on both beams, different from what I thought before. Maybe this is because the tunes are so close.

I got them from the OP log, 20:08 at 6-4-2010. Warning: the log says: Frequency axis slightly off :? .
20100406200908.png
B1
20100406200908.png (128.98 KiB) Viewed 4925 times
20100406200918.png
B2
20100406200918.png (142.27 KiB) Viewed 4925 times
Happy hunting!
- Tau

pixelmasseuse
Posts: 29
Joined: Tue Apr 06, 2010 9:25 pm

Re: More news

Post by pixelmasseuse » Wed Apr 07, 2010 3:01 am

Tau wrote:The hump scan they did today gives a lot of new info for wannabee hump hunters.
hmmm. From the graphs, it looks to me like the interference has two main components. 1. a constant and pretty intense peak at .3 Frev, and 2. A varying component that looks like a mirror of the tune.

Are they sure that this interference is actually manifest as beam oscillations (i.e. not just a measurement artifact)? If so, it seems to me that the .3 Frev component would be causing oscillations at a much higher amplitude than the tune, and it would be more than just an annoyance.

User avatar
Tau
LHCPortal Guru
Posts: 168
Joined: Tue Jan 05, 2010 10:16 pm
Location: Heemskerk, Netherlands

Re: Let's go hump hunting!

Post by Tau » Wed Apr 07, 2010 8:28 am

The "intense peak" is the tune: it is supposed to be there. Normally the B1 and B2 tunes are further apart, but this time they are exceptionally close.
The vague moving lines (note time moves upwards) are the hump.
- Tau

Kasuha
Posts: 570
Joined: Tue Dec 15, 2009 1:22 pm

Re: Let's go hump hunting!

Post by Kasuha » Wed Apr 07, 2010 10:36 am

Apparently both beams have the hump then, not just one. B1's hump is just weaker. It still suggests it is a local source though.
Rather than "main hump" at 0.15~0.2 frev I guess the real source might be near the line at 0.33~0.35 frev, it appears to be better defined than the hump itself. The one at 0.35~0.4 seems to just be a harmonic.

I wonder if it can't have something to do with beam synchronization - beam tracks must be kept equally long to great precision to prevent beams going off sync. This fine tuning is probably done by altering the beam trajectory slightly and it sure has some reaction time...

pixelmasseuse
Posts: 29
Joined: Tue Apr 06, 2010 9:25 pm

Re: Let's go hump hunting!

Post by pixelmasseuse » Wed Apr 07, 2010 2:08 pm

Tau wrote:The "intense peak" is the tune: it is supposed to be there.
D'oh! thanks for the clarification.

It does look like the signal at .33-.35 Frev is the fundamental since it covers the smallest Hz, .175-.21 Frev is the first harmonic (2x frequency spread in Hz), and .35-.42 Frev is the 2nd harmonic (3x frequency spread in Hz)... must be some aliasing going on.

Kasuha
Posts: 570
Joined: Tue Dec 15, 2009 1:22 pm

Re: Let's go hump hunting!

Post by Kasuha » Wed Apr 07, 2010 3:57 pm

On second thought...
The 0.33 line is different from the 0.18 one. It raises almost constantly since beginning while 0.18 starts raising at time 2e5. Also 0.33 is back at base line earlier than 0.18. They are similar but still different.

User avatar
Tau
LHCPortal Guru
Posts: 168
Joined: Tue Jan 05, 2010 10:16 pm
Location: Heemskerk, Netherlands

Re: Let's go hump hunting!

Post by Tau » Wed Apr 07, 2010 6:51 pm

I looked at all lines very carefully, especially at the vagueness, and this is my conclusion:
Main signal: the .33 line (the very thin one)
Second harmonic is .68, which is shown at .18 and its reflection is shown at .32, under the hump.
Third harmonic vanishes in the dark blue (would be at 1.01)
Fourth harmonic, 1.36, which is very visible at .36
Fifth harmonic would be at .69, visible at .19, and could explain the sharp corner near 5.5e5 ms.
This would explain almost everything you can see in the graph.
So the signal has quite a few visible overtones, which probably means it actually is 11.2 * .33 = 3.8 kHz.
- Tau

pixelmasseuse
Posts: 29
Joined: Tue Apr 06, 2010 9:25 pm

Re: Let's go hump hunting!

Post by pixelmasseuse » Wed Apr 07, 2010 7:07 pm

Anyone know what is the sampling rate of the data acquisition?

josch222
Posts: 78
Joined: Sat Dec 12, 2009 4:10 pm

Re: Let's go hump hunting!

Post by josch222 » Wed Apr 07, 2010 7:24 pm

I think I'm not understanding this completely.
So here is what I understand so far and some questions, please tell
me if I'm right or wrong:
The red lines have to be there at a fixed place (but not necessarily at 0.3).
The other frequencies shouldn't be there at all, or is only the variation (jitter, wobble) a problem?

How I have to imagine the beam shape over the circumference?
From what I know now, there is an intentional deviation from a circle (leaving aside that the LHC
is not a circle anyway) which is needed for a somewhat self stabilizing orbit.
That is called "Tune". How do I have to imagine the shape of this tune?
Is it residential? (I think it must be, otherwise there would be problems with colimators etc.)
Would be "srcewing" around the mechanical axis in an odd number of turns the right picture?

Now to the "hump":
Do I have to think of it as residential too, only changing its amplitude?
Or is it circulating along the beam (if so, clockwise or anticlockwise :rolleyes: ),
or is it moving chaotic (back and forth or appearing and dissapearing here and there)?


What I find interesting in the FFTs is the fuzz to lower f of the tune which follows the shape
of the hump frequencies to higher fs. It looks somewhat like excitation of a resonance from noise.

oxodoes
Posts: 66
Joined: Mon Mar 01, 2010 10:26 pm
Contact:

Re: Let's go hump hunting!

Post by oxodoes » Wed Apr 07, 2010 8:33 pm

Josch222: You might want to read this thread first: http://lhcportal.com/Forum/viewtopic.php?f=4&t=219

The shape of the beam is an ellipse of changing orientation and size that is centered around the orbit, which is not necessarily the mechanical axis.

Kasuha
Posts: 570
Joined: Tue Dec 15, 2009 1:22 pm

Re: Let's go hump hunting!

Post by Kasuha » Wed Apr 07, 2010 8:49 pm

pixelmasseuse wrote:Anyone know what is the sampling rate of the data acquisition?
11245 Hz I believe - they simply measure position of the beam every time a bunch passes the BLM.

RocketManKSC
Posts: 40
Joined: Mon Mar 29, 2010 10:09 pm
Location: Titusville Florida

Re: Let's go hump hunting!

Post by RocketManKSC » Sun Apr 11, 2010 12:11 pm

It appears to me that the hump is modulated by some thermal effect. It looks like a heater/cooler come on at about 500,000 ms and off at about 700,000 ms.
If we had, raw data for some of the longer runs you could build a plot of hump Frequency vs. Time overlaying each cycle. If you could match this to some number of heaters/coolers, you could change their set point (cycle time) one at a time to attempt to narrow the search.

RocketManKSC
Posts: 40
Joined: Mon Mar 29, 2010 10:09 pm
Location: Titusville Florida

Re: Let's go hump hunting!

Post by RocketManKSC » Sun Apr 11, 2010 12:18 pm

Note: This problem has some similarity to the (Engine Cutoff) ECO Sensors we had so much troubles with last year. Turned Out to be the connector at low temp

User avatar
chriwi
LHCPortal Guru
Posts: 403
Joined: Fri Nov 20, 2009 1:19 pm
Location: Stuttgart Germany
Contact:

Re: Let's go hump hunting!

Post by chriwi » Fri Apr 16, 2010 6:47 am

Nice comment last night in the e-log: "looks like we're being humped" :lol:
bye

chriwi

Post Reply