Let's go hump hunting!

The place to discuss particle physics
User avatar
Tau
LHCPortal Guru
Posts: 168
Joined: Tue Jan 05, 2010 10:16 pm
Location: Heemskerk, Netherlands

Let's go hump hunting!

Post by Tau » Fri Apr 02, 2010 7:49 am

In a brief conversation with PCatom I found out that the hump is still bothering the operators, and that they're working hard to find it. :?

My idea was to use our collective knowledge to help the CERN people to "hunt the hump".
So, if you have an idea on what the hump could be, on how the cause of the hump could be found, please join in!

We start with what we already know about the hump:
  • The hump is so called because it causes a hump space on the tune graph, which is a FFT transform of the beam oscillations;
    mainly on the vertical one.
    In plain English, this means something is jiggling the beam up and down.
  • If we know the cause of the hump, it is probably not hard to reduce the hump (e.g. by shielding something).
  • The hump decreases the quality of the beams (increasing lifetimes), so we really want to get rid of it.
  • Collimator settings are complicated by the hump, but they do not cause the hump.
  • It must be something that has a magnetic field to move protons (a motor, for example).
  • It is a "noisy" sound, not a clear tone (otherwise it would be a peak, not a hump).
  • Its frequency varies from 2.6kHz to 3.7kHz (that is .23 to .33 times the rotation frequency of the beam of 11.2 kHz).
  • The frequency moves up and down over this range every 7-10 minutes (depending on something).
  • It is still there if SPS is off.
  • In theory, the frequency could be an integral multiple of 11.2 kHz higher (but they think that's unlikely).
If you have an external account, you can find some detailed hump information in the OP elogbook: Personally, my best idea so far is to send someone down the tunnel and let him listen for a hissing sound; and it is relatively easy to check if it is right by comparing it to the hump.

If somebody has an idea, please post! Maybe together we can help them with finding this nasty problem. :wave:
Last edited by Tau on Fri Apr 02, 2010 9:10 am, edited 1 time in total.
- Tau

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

Re: Let's go hump hunting!

Post by Kasuha » Fri Apr 02, 2010 8:06 am

Tau wrote:Personally, my best idea so far is to send someone down the tunnel and let him listen for something that sounds right; it might be recognizable from the description, and it is relatively easy to check if it is right.
Maybe install some portable audio recorders in the tunnel during access (one each 500 meters or so), run some physics, then collect them back, analyze and compare with hump data? Or if there are some intercom devices present, just use them to get audio data during physics run?

My bet would be some cryo pump motor but I guess this got already ruled out among the most obvious causes.

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 » Fri Apr 02, 2010 9:15 am

Good idea, maybe TIM http://www.cernlove.org/blog/2010/01/monorails/ has a microphone on it already.
- Tau

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

Re: Let's go hump hunting!

Post by Kasuha » Fri Apr 02, 2010 9:23 am

It would definitely help to know what was already ruled out. And maybe also how strong the source has to be. Is there e.g. chance that it's caused by something on surface? I can see LHC sits under some quite inhabited areas.

Shadowdraxx
LHCPortal Guru
Posts: 184
Joined: Fri Dec 04, 2009 1:03 am

Re: Let's go hump hunting!

Post by Shadowdraxx » Fri Apr 02, 2010 9:37 am

Kasuha wrote:It would definitely help to know what was already ruled out. And maybe also how strong the source has to be. Is there e.g. chance that it's caused by something on surface? I can see LHC sits under some quite inhabited areas.
Indeed there are alot of Industrial units and overhead electrical pylons that cross paths with the ring.

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

Re: Let's go hump hunting!

Post by josch222 » Fri Apr 02, 2010 10:36 am

Hi,

have they considered ELF efects (Shumann resonances
http://en.wikipedia.org/wiki/Schumann_resonances, U-Boat comm.)
or LF sources (and all their possible mixing products)?

Maybe it helps to place some sensitive measurement equipment some km away
from all the noise generated by CERN itself and log a few days of data in a broad
spectrum to rule out or detect an external source.

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

Re: Let's go hump hunting!

Post by josch222 » Fri Apr 02, 2010 1:54 pm

I searched the net because I knew there are people who are observing ELF and LF.
I'm pretty sure that there is already data, at least at the military but it would be a little
bit difficult to get it. I did not found any public research institution or the like yet,
but this may be helpful:
http://lwca.org/mb/index.html
Another site http://www.vlf.it/
If there is any chance the cause for the hump is external we could ask them if
anybody has logged some larger signal during the occurence of the hump or if
someone is willing to share the data. The LHC-crew could then look for correlations
themselves.

Here is a site with live streaming audio of VLF signals, may browsers don't play them,
does anybody know what plugin or player is needed?
http://67.207.143.181/

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

Remember the facts

Post by Tau » Fri Apr 02, 2010 4:10 pm

Although I think we mustn't immediately reject strange proposals, I find it unlikely that the ELF or LF has the behaviour of the hump: 2.6 to 3.7 kHz, slowly sweeping in frequency between the two in a time of 7 to 10 minutes.
Also, I find it hard to believe the source would come from so far away, but as any true scientist, I will not say it is impossible.
- Tau

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

Re: Let's go hump hunting!

Post by Kasuha » Fri Apr 02, 2010 5:47 pm

Out of wild ideas I had about it:

1/ Sound is currently usually sampled at 48 kHz, beam has frequency of 11,245 Hz - difference between 4 x Beam frequency and sound sampling is 3050 Hz which is close to hump frequency. If there is a loudspeaker somewhere in the tunnel "playing" noise sampled at 48 kHz it could be causing that - and slight changes to noise frequency profile might be causing hump shifts.
Probably stupid, who'd install such loudspeaker there?

2/ Frequency and shifting might coincide with small water power plant on surface (turbine frequency)
There's big lake above but I didn't find any power plant.

3/ There is a lot of electronics in the tunnel, these are usually powered by switched-mode power supplies - these usually oscillate at tens of kHz and the frequency is changed based on conditions. Again, there might be interference between beam frequency and the power supply frequency.
I have no clue what kind of power supply is used in LHC tunnels, I'd rather guess there's some kind of centralized powering at safe distance.

Using more bunches per beam and analyzing frequency characteristics (phase shift) separately for each bunch, if possible, might help locate the hump source too.

JGLambourne
Posts: 10
Joined: Sun Dec 06, 2009 6:52 pm

Re: Let's go hump hunting!

Post by JGLambourne » Fri Apr 02, 2010 7:15 pm

Tau wrote:Personally, my best idea so far is to send someone down the tunnel and let him listen for a hissing sound
I don't think these frequencies are high enough to be a hissing sound. From http://www.phy.mtu.edu/~suits/notefreqs.html we see that 2.6 to 3.7 kHz corresponds to E and A#, 3 octaves above middle C. That's still on a piano keyboard, although way up at the high end.

If the electromagnetic interference were also accompanied by a sound of the same frequency (which is not at all certain) then it would be a high pitched screeching.

My first guess would be a turbo pump on it's way out (although it sounds like cern have already investigated that).

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

Re: Remember the facts

Post by josch222 » Fri Apr 02, 2010 7:39 pm

Tau wrote:Although I think we mustn't immediately reject strange proposals, I find it unlikely that the ELF or LF has the behaviour of the hump: 2.6 to 3.7 kHz, slowly sweeping in frequency between the two in a time of 7 to 10 minutes.
Also, I find it hard to believe the source would come from so far away, but as any true scientist, I will not say it is impossible.
Why should this frequencies and the drift not be in the LF spectrum, there are a lot
of motor controls in industry (ok that has to be nearby to have any effect).
If you look at spectra taken by hobbyists you sometimes see such things,
besides whistlers and chirps (atmospheric effects). btw they use simple
PCs with sound-cards, a magnetic antenna and maybe a simple pre-amp.
And don't forget: The LHC can be seen as a huge LF-Antenna itself.
On the other hand: I'm listening to a LF live-stream from northern Italy for
some time now, no sound in the audio spectrum that would directly fit
the described frequency and drift so far.

In the end it does not matter if the source is nearby or far away, place the right equipment in the tunnel and outside and take data. Use multiple receivers with synchronized and accurate high time resolution and you may be able to locate the source, if it is really a magnetic field.

The idea with microphones is great because it may be really mechanical vibration,
so it would be reasonable to place microphones for structure-borne sound at the mounting posts of some magnets.

I would install both things, log a few days and then analyze the data.

Analyzing what is on the mains (besides 50Hz ;-) would be interesting too, but I
believe they do this already.
Switched power supplies can be a real PITA especially when you have such a large
number and I don't think they are all synchronized. This is always a nice source for
beat frequencies. But such beat should be seen with LF receivers too.

User avatar
morgad
Posts: 48
Joined: Sat Nov 21, 2009 8:43 am
Location: Devon, UK
Contact:

Re: Let's go hump hunting!

Post by morgad » Fri Apr 02, 2010 8:19 pm

if looking for a mechanical vibration in the tunnel, would they be better to bolt some geo-phones to the walls/floor ?

Dave

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

Re: Let's go hump hunting!

Post by josch222 » Fri Apr 02, 2010 9:31 pm

morgad wrote:if looking for a mechanical vibration in the tunnel, would they be better to bolt some geo-phones to the walls/floor ?

Dave
It depends:
If you are sure it is coupled through the floor then yes.
But it may be something that is generated in the magnets itself, for example in pipes vibration can occur when there is liquid or gas flow. And there are some pipes in the
magnets. But either way, in the end it has to act on the beam. So to confirm or rule out
a mechanical source the first attempt should be a measurement as near as possible to the
beam itself imho.

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

Re: Remember the facts

Post by Kasuha » Fri Apr 02, 2010 9:46 pm

josch222 wrote:(besides 50Hz ;-)
The truth is, there is not 50 Hz in the mains, the frequency is slightly changing throughout the day. See here for instance. But I don't see how these slight changes might cause the hump, the interference by far cannot go to 3.5 kHz IMO.

JNW
Posts: 42
Joined: Wed Dec 02, 2009 4:48 pm

Re: Let's go hump hunting!

Post by JNW » Sat Apr 03, 2010 4:22 am

Electrical interference can be caused by an alternative ground path.

Here's an example. Assume we're in the US where many electrical outlets have a 120VAC hot wire and the return is near ground potential. If you plug in an appliance that draws current, then the current in the hot wire and the current in the return wire will be the same, except in opposite directions. There will be some magnetic field very close to the wires, but at a distance the magnetic fields from the two wires will exactly cancel.

However if the appliance is also grounded by another route, and this ground is shorted to the return wire inside the appliance, then some of the return current will flow through this alternative path. The currents in the hot and return wires no longer cancel, and so there will be a net magnetic field, even some distance away. The alternative ground path will also have an unbalanced current, and will also generate a magnetic field. The appliance will work fine, though.

If some wires in the LHC are carrying current with the hump frequency, then a wiring error like this could cause magnetic interference.

Has anyone jogged around the LHC tunnel with a detector while the LHC is running (but with no beam, of course)?

Post Reply