Detector Watch S6 V2

ML1
ML1
Joined: 20 Feb 05
Posts: 347
Credit: 86308057
RAC: 599

RE: RE: ... Or, how about

Message 95205 in response to message 95204

Quote:
Quote:
... Or, how about in the ocean?...

4 km would be possible, but you'd have to have a floating observatory moored in pretty deep ocean. Just think of the fun and games they'd have trying to keep everything rigid and locked-in to within a fraction of a wavelength, especially for the two surface arms.


Don't use mooring lines! Use submerged units and station keeping. It could even be used as a test-bed for the LISA technology.

I'd guess that the station keeping would be no more a nightmare than the presently used anti-seismic supports for the land-based LIGOs...

Also, they wouldn't have problems with bunnies bouncing (thumping) around, or with ocean breakers pounding the shoreline, or with clanking railways. However, access and supplies would be more of a problem...

Regards,
Martin

See new freedom: Mageia Linux
Take a look for yourself: Linux Format
The Future is what We all make IT (GPLv3)

ML1
ML1
Joined: 20 Feb 05
Posts: 347
Credit: 86308057
RAC: 599

RE: RE: ... Or, how about

Message 95206 in response to message 95204

Quote:
Quote:
... Or, how about in the ocean?...

4 km would be possible, but you'd have to have a floating observatory moored in pretty deep ocean.


Would the 4km deeper (and slightly slower in time) into Earth's gravity well be great enough to be significant and a problem?

Regards,
Martin

See new freedom: Mageia Linux
Take a look for yourself: Linux Format
The Future is what We all make IT (GPLv3)

Bikeman (Heinz-Bernd Eggenstein)
Bikeman (Heinz-...
Moderator
Joined: 28 Aug 06
Posts: 3522
Credit: 686043288
RAC: 582127

RE: RE: RE: ... Or, how

Message 95207 in response to message 95206

Quote:
Quote:
Quote:
... Or, how about in the ocean?...

4 km would be possible, but you'd have to have a floating observatory moored in pretty deep ocean.

Would the 4km deeper (and slightly slower in time) into Earth's gravity well be great enough to be significant and a problem?

Regards,
Martin

I guess as it doesn't change with time, it would be possible to sort this out by correction.

But building the stuff in water....come on.... . (I think there is a saying for movie directors that there are three ways to get trouble on the set: work with animals, work with kids, or work with water!)

I guess a floating or submerged detector could detect a lot of interesting things like whales mating a few 1000 miles away (acoustic interference) or whales mating nearby (by a gravity gradient), but GWs??

CU
H-B

Mike Hewson
Mike Hewson
Moderator
Joined: 1 Dec 05
Posts: 6534
Credit: 284710168
RAC: 110462

RE: Would the 4km deeper

Message 95208 in response to message 95206

Quote:
Would the 4km deeper (and slightly slower in time) into Earth's gravity well be great enough to be significant and a problem?


Well spotted! Indeed there is a time effect over that scale to be accounted for. One of the initial terrestrial tests of gravitational redshift was done over ~ 100ft tower. For the current IFO's I'm pretty sure the 'fall' over the 4km arms has been engineered to insignificant levels. But bear in mind that over such a distance there will be an actual curvature to the Earth - or if you like the direction to the 'centre of the Earth' will change slightly from corner to end stations. This is not as trivial as it might seem, Gravity Probe B had to correct for the different placement of the 4 tests masses within the craft.

Cheers, Mike.

I have made this letter longer than usual because I lack the time to make it shorter ...

... and my other CPU is a Ryzen 5950X :-) Blaise Pascal

ML1
ML1
Joined: 20 Feb 05
Posts: 347
Credit: 86308057
RAC: 599

RE: RE: Would the 4km

Message 95209 in response to message 95208

Quote:
Quote:
Would the 4km deeper (and slightly slower in time) into Earth's gravity well be great enough to be significant and a problem?

Well spotted! Indeed there is a time effect over that scale to be accounted for. One of the initial terrestrial tests of gravitational redshift was done over ~ 100ft tower. For the current IFO's I'm pretty sure the 'fall' over the 4km arms has been engineered to insignificant levels. But bear in mind that over such a distance there will be an actual curvature to the Earth - or if you like the direction to the 'centre of the Earth' will change slightly from corner to end stations. This is not as trivial as it might seem, Gravity Probe B had to correct for the different placement of the 4 tests masses within the craft.


Mmmm... That finely measured eh?!

For the ocean based LIGO, let's say name it OLIGO, I would guess that the slower time as you go deeper into the ocean just means that you adjust the length to be slightly shorter to 'see' the same light round-trip-time as seen for the near surface 4km arms.

So... Use isolated nodes as with LISA or replicate the LIGO vacuum pipes?

I guess that changing refractive index due to temperature variations would play havoc with trying to measure distance and for maintaining alignment. Hence, vacuum pipes required. Shame that will bump up the costs somewhat...

So... For OLIGO take advantage of contriving neutral buoyancy and use station keeping for all the segments. Hence the tubes for the arms can be made to be made as lightweight as necessary to survive just the water pressure.

That just leaves how to power and supply the thing...

Perhaps LISA is more cost effective than an OLIGO!

Then again... Before all the ice melts on us, there's 2km or more depth of ice in Antarctica for a 3 arm ice-LIGO :-)

Keep searchin',
Martin

See new freedom: Mageia Linux
Take a look for yourself: Linux Format
The Future is what We all make IT (GPLv3)

Mike Hewson
Mike Hewson
Moderator
Joined: 1 Dec 05
Posts: 6534
Credit: 284710168
RAC: 110462

RE: Mmmm... That finely

Message 95210 in response to message 95209

Quote:
Mmmm... That finely measured eh?!


Oh, yeah.

The Gravity Probe B idea is to have a test mass in free fall around the Earth, but with a craft wrapped around it to ward off other influences. Fly the craft around the test mass while not quite touching it - using thrusters and what-not. Hence 'drag free test mass'. For redundancy put 4 in the craft. But they can't all be in exactly the same place can they? Strictly speaking each of the 4 masses experiences a slightly different force from the Earth, because of their non-identical positions at any given time, and hence have slightly different orbits when left alone. So the craft can never exactly follow the orbits of all four simultaneously.

This is generically called 'tidal' effects. Forget about water and the moon in this discussion.

Suppose you have two masses orbiting, but both aligned along a line of radius outwards from a third much larger central body. The nearer one has a greater force toward the third and the far one a lesser force ( direction of force same but magnitudes different ). This is inverse square, or even whatever applies with GR adjustment at any tremendously high field strengths. In any case this will result in a relative movement between these two orbiting bodies - they will gradually drift apart along the radial line. This also explains the ripping apart of that hapless textbook astronaut who recurrently descends feet first into black holes. :-)

Now suppose you have two masses orbiting, with each at the same radial distance from the central body ( ie. one slightly 'leading' the other in the direction of orbit ). The magnitude of the central force is the same for each, but now at slightly different directions - the vectors for each converge to the core of the central body ( the radial vectors from each orbiting mass to the central body are non-parallel ). From the position of one orbiting body, the direction of centripetal force for the other has a slight component towards it. Thus in time they will gradually drift together.

For two orbiting masses within both different radial directions and distances there is a combination of the above. And with four ..... :-)

Quote:
For the ocean based LIGO, let's say name it OLIGO, I would guess that the slower time as you go deeper into the ocean just means that you adjust the length to be slightly shorter to 'see' the same light round-trip-time as seen for the near surface 4km arms.


Well suppose you manufacture two identical arms ( 4km each ) and use one as the 'horizontal' and the other as the 'vertical'. Then from the point of view of the corner station ( where you will be comparing relative phases of the arm round trips ) the photons traveling the up/down arm will have fewer total phase cycles for an 8km round trip than the across/back arm. But that's still OK as once lock is achieved for a given setup ( when no GW's are passing ) then GW's will later affect that.

[ Note that the test masses themselves need to hang 'vertically' regardless of the overall orientation of the arm. Including the test mass/mirror at the lower end of the vertical arm. But a few extra mirrors can sort that out - like a periscope arrangement really. ]

Quote:

So... Use isolated nodes as with LISA or replicate the LIGO vacuum pipes?

I guess that changing refractive index due to temperature variations would play havoc with trying to measure distance and for maintaining alignment. Hence, vacuum pipes required. Shame that will bump up the costs somewhat...


Yeah, vacuum is the go. We're after 10^(-21) order influences so you want to clear the deck of all that you possibly can. As this is not primarily a study of materials science ( although we need to know alot of it ). :-)

Quote:
So... For OLIGO take advantage of contriving neutral buoyancy and use station keeping for all the segments. Hence the tubes for the arms can be made to be made as lightweight as necessary to survive just the water pressure.


Hmmmm. Atmospheric pressure at sea level can support a column of water some 34 or so feet high ( if a vacuum is above the water ). How many lots of 34 feet is there in 4km? About 400. Multiply that number by sea level atmospheric pressure and you have a specification for the required strength of the container for our vacuum .... ( 400 x 14.7 psi )

Quote:
That just leaves how to power and supply the thing...


That's looking like the easiest task now! :-)

Quote:
Perhaps LISA is more cost effective than an OLIGO!


And it will be looking at a different GW frequency range - far lower - than LIGO. In essence it will be sensing activities of a far larger masses and sizes. Right up to cosmic! :-)

Quote:
Then again... Before all the ice melts on us, there's 2km or more depth of ice in Antarctica for a 3 arm ice-LIGO :-)


Let's put in a funding proposal. See if we can confuse it with that other meaning of 'GW' and we're in! :-) :-)

Cheers, Mike.

( edit ) If you go up 4km in a tower, you lose the pressure differential problem b/w the outside air and the inner vacuum - but also any significant buoyancy to hold the tower up. It's a cruel world. :-)

I have made this letter longer than usual because I lack the time to make it shorter ...

... and my other CPU is a Ryzen 5950X :-) Blaise Pascal

Mike Hewson
Mike Hewson
Moderator
Joined: 1 Dec 05
Posts: 6534
Credit: 284710168
RAC: 110462

A review of the last few

A review of the last few weeks.

Livingston : Here's views of the wave heights banging the western Atlantic :

Plus they've had lightning, thunder, heavy winds and rain. Snow also. And the standard earthquakes, planes, trains and automobiles. So only the briefest of locks, if any. Many shifts with 0% duty cycle. Note that one needs 300+ continuous seconds of science mode to achieve a data segment, and hence be included in a duty cycle calculation ( or get passed downstream for analysis ).

I'm not sure if this walrus sans tusks reflects the Louisiana weather, the IFO operators' morale, or what :

Hanford : Here's a report on general performance of H1 for this science run uptil 01/12/09.

Quote:

Thanksgiving Weekend (11/25-11/30):
34 Science Segments
381365 seconds of Science Data
518400 seconds in total for the weekend
74% duty cycle
11216.62 seconds per science segment
3.12 hours per science segment

Commissioning Break was 9/1-9/18, 9/24 was first lock of S6b.
ALL OF S6a (Before Commissioning Break), 7/8-8/31:
409 Science Segments
S6a total science seconds 2372213.00
S6a total seconds 4752000.00
S6a mean length of science segment 5800 seconds
S6a Duty Cycle 49.92%
S6a Mean Range 9.6005 Mpc including non-Science time, i.e. overall.

ALL OF S6b (9/24-11/30)
326 Science Segments
S6b total science seconds 2809617.00
s6b total seconds 5875200.00
S6b mean length of science segment 8618 seconds
s6b duty cycle 47.82%
S6b Mean Range 8.7433 Mpc including non-Science time, overall.

Science Segments based on this.

Virgo is reporting 80.6% duty cycle for VSR2.

The difference in range between S6a and S6b is 91.0713% (S6b/S6a) the length of range overall.

Corresponding to volume, if you cube that value you get 75.5344% the volume of range overall.

Science segments in S6b are 49% longer than science segments in S6a.


So that roughly means longer segments now, but shorter range and thus volume, on an overall slightly less duty cycle. Adding up the duty cycles for the 37 shifts from then ( 01/12/09 ) until now gives 62.84% - spoilt really by a half dozen very difficult shifts ( very windy ). However the ranges were well above 10Mpc. There were segments with :

Quote:
We had a fair amount of triple coincident data today!


Note that this is not the H1/H2/L1 triple coincidence ( all taking data at the same time ) of old but H1/L1/Virgo ! Which is great, eh!

Let's think about this duty cycle/co-incidence thing.

If I get a 'signal' ( parentheses indicating it is a candidate that has passed later analysis with sufficient signal/noise plus other tests ) at one IFO only, then how sure are we of this as a 'real' one? Probably not very.

If I get a signal at two IFO's, and pretty close timewise ( comparable to direct light travel time b/w the IFO's ), then we'd be more confident of the fact of a GW detection. [ except the previously used H1/H2 pair could only at best see the same waveform - but H2 responding at half strength ]

Now if you get three ( well separated ) IFO's with signals near co-incident, we would be real happy. This could conceivably yield a firm direction to source ( two would confine to a band on the sky ).

Now four IFO's, ideally, yields direction and range to source. Patience ... :-)

For the current three IFO set then duty cycles roughly multiply to give the expected fraction of triple co-incidence time. So if all three were 80%, say, then triple time = 0.8 x 0.8 x 0.8 = 0.512 or 51.2 %. Which alas doesn't nearly sound as good. However you'd still get 64% ( 0.8 x 0.8 ) expected double time for the IFO's taken as pairs. As (0.95)^3 ~ 86% then that would be real nice - you can see why banks make money using compound rather than say simple interest.

There's been a large animal spotted nearby too :

maybe related to 'the bizarre nightly seismic activity in Hanford'. Who can say?

Cheers, Mike.

I have made this letter longer than usual because I lack the time to make it shorter ...

... and my other CPU is a Ryzen 5950X :-) Blaise Pascal

Mike Hewson
Mike Hewson
Moderator
Joined: 1 Dec 05
Posts: 6534
Credit: 284710168
RAC: 110462

Livingston has had a quieter

Livingston has had a quieter time and better success. Less external ground movement overall, but also improvement in isolation. Here's one aspect reported :

Quote:

Expected performance on PRC_CTRL from Y feed forward
Last week we implemented the correction to the STS-X feed forward to RM HEPI. That improved the isolation at useism frequency by a factor of 2-3 and consequently improved the PRC angular motion and the interferometer noises and stability.

The attached plot shows the potential improvement if we also implement the STS-Y feed forward. The data are from the day time full interferometer lock (the filter for STS-X was calculated for data in PRM configuration) with useism between 50 and 95% level. Currently we don't have the hook to feed it to the X of the RM HEPI. But we can try some other place like Y direction of the ITMY HEPI. Although now is the time to think how we are going to do the arms.


This refers to the Power Recycling Cavity, so only a portion of the interferometer ( to be extended to the arms ). The idea is to alter the control loops accordingly in feed-forward ( basically correct in advance/expectation of movement ) and dampen energy of motion using the Hydraulic External Pre-Isolation. Thus the red curve is below the blue one showing less vibrational energy going to the interior. This is an operator's comment on it's use :

Quote:
Microseismic coming down from the last shift. When it came down to a reasonable level, detector was able to lock with use of the new filter.


One shift was 100% and several close to it. Check this out for Hanford - several shifts with 100% science mode :

Quote:
probably due to the snowy roads, so very little traffic compared to usual.


Quote:
Snow and freezing rain over the past day/night has made driving difficult, though not impossible. I myself max'ed out at 20 mph on Hwy. 10 last night.


May we have more of these, please ( today's log ):

Quote:
Duration: 68412 seconds, just over 19 hours, ending 24-hour rally. Goals unrealized, hopes dashed, but Xmas will be our next best shot...


and some of these too :

Quote:
Triple Coincidence Trigger Yielded Call To Control Room
--Received a call from Amber [Data Analyst] regarding Triple Coincidence Trigger (she called x203...should call x202!). Whenever an Operator gets a call like this, one should give a brief status to the ifo's Science Mode status (i.e. everything good or bad). The Data Analyst will then request the SWIFT to scan for any events around our triple-c event.


which, if I'm not mistaken, refers to closer attention/examination of the records to periods of time where Hanford/Livingston/Virgo were all simultaneously taking good science data AND each got a near-simultaneous reception of something AND there may have been a gamma ray burst event detected via satellite. I get goosebumps on my goosebumps just thinking of that .... :-) :-)

As regards Piezo-Electric External Pre-Isolator, an active seismic damping system :

Quote:

PEPI Working Strongly Even In Tumultuous μSeismic Times
.......
Conclusion:

Though we aren't reaching 17 Mpc at all (remember when 18 Mpc was our goal?), we are:

* No longer losing lock due to high microseism
* Maintaining decent range at ~16Mpc comfortably in all conditions
* Not oscillating in range due to microseism, and thus no longer seeing DMT DQ flags for high range stdev in Omega Scans
* Duty cycle has been ~70% with commissioning time.
* Run at 14W for significant time on a weekday.
* Robert Schofield did a great job on this one.


I won't burden with the gory detail here, but if you want a peek at the sort of matrix/alignment considerations mentioned earlier ( but referring to the Output Mode Cleaner mirrors, not the IFO as a whole ) then see here. Here's is how adequate clamping is defined :

Quote:
ALL mounts must be FIRMLY attached to the table, i.e. if there is 1/2 a turn left in the screw, then it's too loose, and will cause us noise problems, alignment problems, and down time.


I just love some of the finer engineering points ( add in a water reservoir to mop-up/reduce propagation of external vibrations! ) :

Quote:
The peak at 55.9 Hz appears to be due to the chilled water pump and, as shown by the blue trace, is huge on the flow meter and on the tubes away from the TCS table. I traced it back along the water line to where it disappeared at the junction where the new rubber hose from the chiller closet connects to the old copper tubes that run the rest of the way to the TCS table. The observation that it is on the copper but not the rubber tubes suggests that we might mitigate it with an expansion bladder near the TCS table.


It's great to see some real progress in reducing the noise budget items. That energy just seeps in, but also note that due to up/down-conversion then deflecting energy entrance at one frequency helps reducing noise at other frequencies as well.

Cheers, Mike.

I have made this letter longer than usual because I lack the time to make it shorter ...

... and my other CPU is a Ryzen 5950X :-) Blaise Pascal

Mike Hewson
Mike Hewson
Moderator
Joined: 1 Dec 05
Posts: 6534
Credit: 284710168
RAC: 110462

RE: I just love some of the

Message 95213 in response to message 95212

Quote:

I just love some of the finer engineering points ( add in a water reservoir to mop-up/reduce propagation of external vibrations! ) :

Quote:
The peak at 55.9 Hz appears to be due to the chilled water pump and, as shown by the blue trace, is huge on the flow meter and on the tubes away from the TCS table. I traced it back along the water line to where it disappeared at the junction where the new rubber hose from the chiller closet connects to the old copper tubes that run the rest of the way to the TCS table. The observation that it is on the copper but not the rubber tubes suggests that we might mitigate it with an expansion bladder near the TCS table.


I'll explain this a bit better. Copper is harder than rubber, which is floppy by comparison. For a given ( small ) force copper will move much less than rubber and also return to it's prior shape better. When you bend/stretch rubber the molecules within move relative to one another and randomise the orderly motion of a sound wave - heat is generated. So when impulses of pressure travel up the core of the rubber tubing it expands and contracts as the pulse passes. In this way the vibration being transmitted through the water, from the chiller pump's action, is converted to heat ( and/or gradually contributes to permanent deformation of the rubber matrix ).

With the copper there is, by comparison, little relative movement within the metal lattice and the impulses are transmitted onwards with much less energy being absorbed. So if you put in a bladder, basically a rubber water balloon, then you make available more rubber to expand and contract - thus soaking up more sound wave energy. Also there will be some extra volume of water in the bladder, which will likewise assist. Remember if you stir a cup of coffee, it gets slightly warmer due to that - so the 'directed' energy of the spoon's movement dissipates as heat.

Cheers, Mike.

I have made this letter longer than usual because I lack the time to make it shorter ...

... and my other CPU is a Ryzen 5950X :-) Blaise Pascal

Mike Hewson
Mike Hewson
Moderator
Joined: 1 Dec 05
Posts: 6534
Credit: 284710168
RAC: 110462

RE: PEPI Working Strongly

Message 95214 in response to message 95212

Quote:

PEPI Working Strongly Even In Tumultuous μSeismic Times
.......
Conclusion:

Though we aren't reaching 17 Mpc at all (remember when 18 Mpc was our goal?), we are:


This, I've just noticed, answers a question asked a few months ago - the Enhanced LIGO inspiral range goal. I'd thought ( perhaps falsely based on other reading ) that it was twice that. Ahh, factors of two .... :-)

Cheers, Mike.

I have made this letter longer than usual because I lack the time to make it shorter ...

... and my other CPU is a Ryzen 5950X :-) Blaise Pascal

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.