Tuesday 31 May 2011

Demolition team

In addition to the author and the local contractors, here are the other members of the SGO team who completed the site testing, demolition and clearing work during May 2011.

Pertti Ylitalo, Toivo Iinatti and Markku Postila.
Saana mountain is in the background.

Monday 30 May 2011

Destroying LOFAR — The last cell

'The year is 2011 C.E. All of the raised test-tile has succumbed to snow loading. All? Not quite! The indomitable Cell-#0 held out, as strong as ever...'

When the test tile collapsed, every cell in the tile was affected in some way. Either the wall collapsed, the innards were crushed, or the polystyrene was either bent or fractured. However, there was one exception.

This was Cell #0, which was located in the opposite far right corner from the location where the snow was being dumped. When we dismantled the tile, it turned out that both the cell walls and cell contents survived unaffected.

As a result, we carefully removed this unit, packed it, and transported it back to SGO. There is can be used as a reference and test cell.

The only cell to remain intact... just before being loaded into the van.

Sunday 29 May 2011

Now that's cute...

While all this work is going on at KAIRA (Finnish LOFAR station FI609), our colleagues to the south have also been busy.

Work at the Onsala Space Observatory has also been progressing well. The first HBA tiles have started arriving for LOFAR station SE607 and work is ready to get under way with their installation. If you haven't seen it yet, we recommend you visit their LOFAR-Sweden web log. Apart from some interesting articles, they have a fantastic collection of great photographs on their Picture Gallery.

This one is particularly sweet. It is not clear from their site who made that sign, but you have to admit it is cute.

Nice work, Onsala. All the best for your HBA build!

Photo: Onsala Space Observatory.

Saturday 28 May 2011

Parhelion revisited

From rainbows to the aurora, our atmosphere is capable of generating some spectacular light effects. A few days ago, we posted a photograph of a parhelion. This is a light effect, caused by the scattering of sunlight from ice crystals in the atmosphere. Now, thanks to a recommendation from Jenny Shipway at the InTech Planetarium, we've found some interesting software that allows anyone to experiment with this phenomenon.

The recommended site was http://www.atoptics.co.uk/ — a site dedicated to atmospheric phenomena. The relevent section for what we saw from the KAIRA site is under 'Ice Halos'. Additionally, there is a link to some Windows software that anyone can download and use. And, that's what we've done here to re-create the effect that was seen at Kilpisjärvi.

The software does need some data. There was the time and date of the photograph (11-May-2011 16:48 UTC), from which we can use another online tool to get the solar elevation (as we didn't measure it at the time!). This gives a solar elevation of 16.9 degrees.

Below is the simulation (top image) and actual photograph (below it). As you can see from the comparison, it does a pretty good job of modelling what we actually saw.

Obviously there are a lot of lens flares within the photograph due to the limitations of the camera (both the lens and the CCD sensor therein) and I could have spent more time tweaking the camera angles to get the alignment correct. Still, it is a good match and indicates that the physics is well understood within the software.

The programme is pretty sophisticated and one can tweak parameters such as the ice crystal shapes and orientations to explore different effects.

Friday 27 May 2011

Destroying LOFAR — Removing the frame

The final stage in clearing the area where the raised test tile was, is to get the frame itself out of the way. This also provided a useful test in that we could experiment with the moving of an entire frame using the 15-tonne digger provided by the local Kilpisjärvi contractor.

Markku Postila and Toivo Iinatti attach the lifting straps.

The frame is moved off the edge of the site.

With the frame completely moved, and ground tile also gone, the site is now fully clear ready for the ground preparation work for the complete build.

Ground works will actually start in a few weeks, giving plenty of time for some surface thawing of the ground, making it easier to work with.

Thursday 26 May 2011

Frame anchor condition

As we saw in our article on the ground tile anchors, they held firmly in the ground and the snow drifts did not affect the translation or tension on them. As it turned out, the anchors holding the raised tile to its frame, and the anchors that in turn hold the frame to the ground fared even better. Here are a few photographs (taken before the tile collapse!) of the anchors holding the frame to the ground lines.

One thing to note is that the anchor lines remain vertical. What this means is that the frame has not shifted in any way. One of our concerns was that wind and snow loading would cause the antenna tile or the support frame to shift (similar to our shift-checks on the ground tile). As it turns out there is no evidence of even the slightest displacement.

Wednesday 25 May 2011

Moving the ground tile — Part 2

Once the tile is lifted, it needs to be carefully driven across to the final storage location. This was done by simply driving the lorry with the HIAB (vehicle-mounted crane) over the final location. This was very risky for the tile. The lorry does not have extra-soft suspension and there is no shock-absorbency in the lifting rig. The move was complicated by the (comparatively) rugged terrain and the fact that the breeze was stiffening.

Still, we managed to move it without losing any components and then set it down onto a prepared pallet which would act as a storage base.

The HBA tile touches down on its storage location.

At this point, the tile remains very fragile, so it needs to be secured right away. A tarpaulin was quickly placed over the tile to contain the parts and provide the inner layer of protection. Then a timber frame was put in place to provide structural stability.

Shoring timber is applied to support the tile.

Once that has been done, then the external covers are applied and the system pegged down. The lids are stacked alongside and are also covered and secured.

Nearing completion. Additional covers are applied and tile lids are stored.

Some additional clearing and securing was then done and now the tile is in a safe state to endure until possible requirement during the main HBA build, or even perhaps the future expansion of the KAIRA system.

Photos: D. McKay-Bukowski

Tuesday 24 May 2011

Moving the ground tile — Part 1

As we know from the various photographs and reports, there are (were!) actually two test tiles at the site. Obviously, our recent series has focused on the destruction of the raised tile.

But what of the ground tile?

We want to move this tile off the main build area. We need to get it out of the way so we can build the main array. It is also an 'empty tile' in that there are no electronics or receiver elements embedded in it.

On the other hand, we want to keep the tile, as it is both useful as a source of parts, but also as a possible 'centre-tile' in the case we upgrade to an international station size for the KAIRA HBA.

Therefore, we decided to move the tile, pack it, and store it on the site.

The first step in this process is to remove the cover of the tile and all the solid polystyrene lids. These are stored separately. This leaves the tile vulnerable, with all the delicate parts exposed, so we need be careful now.

This is the tile with the cover removed. It clearly shows the pristine polystyrene structure.

Once the tile is ready, it can be refolded. So often, we hear about unfolding the tiles, as reported on other web sites about other LOFAR stations. However, as far as we know, this is the first time a tile has been re-folded in the field. (We'll ignore the manufacturing set-up which has the facility to fold the tiles as they are built.)

Then we prepare to lift the tile.

It is at this point that we are ready to lift the tile. Normally during unfolding, there are batons, wrapping and straps. As we are refolding, there is none of this. We didn't even have a correct centre beam, as the main LOFAR tooling is currently in use in Sweden on the SE607 build.

HBA tile being lifted

Once lifted, the tile can then be refolded and moved. We did not have the possibility to load the tile onto a pallet, nor do we have the straps or batons to secure the individual components. They are held in place at this 'unexpected' angle by gravity and gentle nudging by the team.

Nervous moments: the tile parts totter precariously in the breeze.

What happened next will be explained in tomorrow's report.

Photos: D.McKay-Bukowski

Monday 23 May 2011

KAIRA and Grímsvötn

As many of our readers will know, Iceland and its spectacular volcanoes have been in the news again. This time it is not Eyjafjallajökull, but the Grímsvötn volcano. Although there have been some recent volcanic activity from Grímsvötn (1998 and 2004), this recent event is much greater. The eruption, which began on 21st May 2011, is reported as being the largest from that particular volcano in over 100 years. There have been many reports about this, including some spectacular photographs, made in the general press.

As might be expected, there is a lot of interest in the event due to the disruption that Eyjafjallajökull caused last year to air traffic. There is also the question as to whether any effect would be noticed from Kilpisjärvi. However, predictions at the moment suggest that the bulk of the ash cloud will move to the north, and should be well clear of the KAIRA site.

That said, it wouldn't take much for the weather pattern to change and for us to see some interesting sky colours due to the scattering of sunlight from ash particles in the upper atmosphere — especially as we are 'midnight sun' season and the sun shines from above the northern horizon at certain times. And, although it is almost certain that there will be no ash falling on the site, our recent snow/ice loading tests indicate that even if there was, our array would be able to cope pretty well. ;-)

Destroying LOFAR — Clear away

Now that the test tile has been destroyed, and the data regarding its structural failure accumulated, we need to clear the debris away so that the site can be prepared for the main build.

Markku Postila pulls fragments from the HBA wreck.

Late into the 'night', the team continue working to get the frame clear.

Sunday 22 May 2011


During the recent site preparation and testing, the KAIRA team was treated to a spectacular parhelion. Parhelia (also known as "mock suns" or "sun dogs") are patches of light that appear 22 degrees to either side of the sun. The effect is caused by ice crystals in the upper atmosphere. The parhelion can be seen at the same altitude as the sun, near the right-hand edge of the image. Unfortunately, the photograph does not do it justice. The view is looking northwest from the LOFAR-FI609 HBA site.

Although in the image there are various 'lens flares' around the sun itself, the parhelion is a real atmospheric phenomenon, and is clearly visible to the naked eye. This is not the first time we've seen this effect over a LOFAR site. In fact, I photographed one from the LOFAR-UK site in Chilbolton last year (LINK).

Saturday 21 May 2011


At one point we left the collapsed tile to go and deal with other parts of the site and then to get some food. On our arrival back, we noticed there was a little visitor to the collapsed HBA tile.

This is a lemming (Lemmus Lemmus). These small furry creatures (about 10cm long) are subnivean animals, meaning that they live at the layer where the snow meets the ground. As the snow has been melting we have noticed a few of their little tunnels at the ice-edges. There are actually a lot of these creatures about and we have seen many now. Although shy, I managed to get close enough to take the above picture, before it jumped over the timber and scurried away over the snow banks.

Lemmings are probably most known from a famous computer game of the same name, where the player had to save these hapless creatures from disaster. Personally, I don't recall a level where you had to fetch them out from collapsing LOFAR HBA tiles.

Friday 20 May 2011

Dummy-tiles on international stations

As we've been discussing the prospects of upgrading KAIRA to an international-station size, this would mean converting the HBA array to full size (KAIRA is already at full size for the LBA and RF-electronics). This would mean incorporating an additional 49 tiles.

This often causes some confusion. There are actually 97 HBA tiles in a full international station. However, only 96 of them are used. Let's consider this tile-and-cable layout digram (from the UK608 station at Chilbolton). The numbered locations 0–96 are actual active tiles.

Even though there is a gap on the plan, a tile is actually put at this location.

This centre tile will not be a working antenna. Instead, it is just a tile. It will have the front end elements (to maintain RF coupling to its neighbours) and will have the exact same outer shell. But the signal that goes into it is terminated immediately and it has no cables.

The main purpose of this dummy tile is to maintain a uniform surface across the array for wind-loading. The 96 is a good round number (in digital processing terms) for the handling of the signal data. However, 96 is not symmetrical when put into an antenna pattern.

Originally, in other LOFAR arrays, the centre tile was simply omitted. But it was found that on the other stations that had already been completed, turbulence cells were being generated as the wind went over the gap. Thus, all new stations are being fitted with a dummy tile in the centre.

In this cable schematic for the HBA field, the tile layout is clearly visible showing the missing 'antenna' at the centre of the array. However, this is where a dummy tile will be placed, thus resulting in a uniformly black plane of tiles on the finished field. And this is precisely what we see in completed systems (such as in aerial photographs).

The last sunset

The season of Midnight Suns starts in Kilpisjärvi on about the 20th May and lasts until the 23rd July. We've now had our last 'horizon' sunset for quite a while!

However, there are a number of prominent mountains about and there sun will continue to disappear for a few hours here and there. However with it always above the horizon the sky will not dim again until the autumn.

Thursday 19 May 2011

Destroying LOFAR — Post Mortem

With the HBA cleared, we can start to analyse what caused the tile to fail and in what way that failure propagated through the structure.

Looking at the split-point in the tile. The cover was not damaged at all.

Here we've peeled back the cover to reveal the
end column. Cell 11 is split down the middle.

Removing half of the lid shows the destruction inside.

The complete mess inside Cell 11.

With the shards removed from the cell content,
the failure of the cell walls can clearly be seen.

A full catalogue of damages was made, along with hundreds of photographs. Although the analysis of the tile failure is mostly a curiosity for the already-deployed LOFAR network, this is a crucial learning opportunity for the proposed EISCAT_3D system.

Wednesday 18 May 2011

Destroying LOFAR — Removing the last of the snow

Although the digger could remove much of the 7.6-tonnes of snow and ice on the destroyed HBA tile, we did not want to risk disturbing the broken pieces and thus confuse the damage caused by the collapse with additional destruction from the clearing process.

Therefore, the remaining few tonnes were done by hand.

Here is the author digging out the area over the collapsed cells.

As we were clearing, there were numerous places where
melt water was starting to stream off the tile edge.

Markku Postilla clears the final few patches of
ice. Saana Mountain is in the background.

Tuesday 17 May 2011

Destroying LOFAR — Starting to remove the snow

Left with a mountain of snow on the collapsed HBA tile, it is now time to clear that snow away again, so that we can dissect the corpse of the deal antenna shell.

The first part of this is to shift the bulk of the ice-mountain. This was done using the 15-tonne digger.

This operation was carried out just as carefully as the initial placement. What we did not want to do was to pack the snow any further, nor generate additional damage by disturbing the now fragile tile pieces with movement from the digger bucket.

Once removed, the snow could be then safely dumped off the edge of the mound which will ultimately be the full HBA site.

EGU trailer

Hey! For those who remember the series we did on the European Geosciences Union General Assembly 2011 there's one last thing to post up.

The EGU has put up their trailer, not only as a look at the scenes from the successful event, but also to announce the dates for next year's event.


Monday 16 May 2011

Frames delivered

It's getting serious now. Shortly after the winter testing was completed and cleared away (although we've still lots to report on this), we had the delivery of all the frames for the High-Band Array (HBA).

In total there are 350 frames (we need 336 for the antennas, plus another 10 or for cable looms, and then some spares).

These are in the middle of the array area at the moment, so they will need to be moved later, but delivery can't be done quite the way we wanted it due to the fact that the ground is extremely rocky in places still, and it is not possible to get the delivery lorry into the exact place we want it.

Destroying LOFAR — Collapse!

Firstly, we'll start this report with a quick apology for all of those who have been sitting clicking 'refresh' on their browsers for the last couple of days. There have been some technical problems with the web log provider, which have caused some delays in posting the next reports on the LOFAR destruction tests.


When we last left our heroes, they have piled over seven-and-a-half tonnes of icy snow on the raised HBA tile. An ominous crack had been heard and the digger engine was shut off and the team started checking the tile, noting some deformation under one of the cells.

Then, we started hearing clicks and creaks. As the gentle movement of Arctic air played over the mound of ice, the creaking would intensify and then subside. It genuinely sounded like being on an old sailing ship, with its wooden hull and masts flexing in the swell and breeze.

And then, the clicks and snaps and pops and creaks started to pick up pace and with a rush of noise that sounded like a tree being felled, there was splintering sound a sudden crash and most of the tile (and the snow on it) dropped down 20cm or so.

Part of the tile edge, showing the crushed tile.

The tile label and one of the anchor points. As the tile
did not fall through completely, no anchors broke.

As can be seen in this last image, the south-west side seems
to have remained partially intact. This is was, of course,
where the pile of snow was the thinnest, and also on the
opposite side of where the digger was dumping the snow.

What needs to be done now is to carefully clear the snow off the tile and then systematically dismantle the remains. What we want to do is look for not only the extent of the damage, but also the failure modes that triggered during the final milliseconds of this tile's life.


Here is a photograph of the lake by the village of Kilpisjärvi, taken on the day the crew arrived (9th May 2011).

One would be tempted to say 'Kilpisjärvi Lake', but that would be tautological, as 'järvi' means 'lake'. The Könkämäeno River flows out of this lake and the waters then join first the Muonio River and then Torne River, before finally flowing into the Gulf of Bothnia.

As you can see, when the photograph was taken, the surface of the lake was still frozen. But it is thinning out in places and has started to fracture, especially around the shoreline.

Sunday 15 May 2011

Kilpisjärvi Biological Station

The KAIRA project is really fortunate in that there is some convenient accommodation nearby which can be used as a base for work on the site. This accommodation is part of the Biological Research Station, which is operated by the University of Helsinki. The station was first established in 1964 and has ever since been used to promote and support Arctic biological research.

The signpost from the main road. Yes, that's part of Saana in the background.

The station belongs to the Faculty of Biological and Environmental science of the University of Helsinki.

However, apart from being very convenient and a delightful and well-serviced place to stay, we actually owe a great deal to the Biological Research Station. The reason for this goes back to August of 2010 when we were still looking for a suitable location for placing the KAIRA project. It was actually in a meeting with Prof. Antero Järvinen (the director of the Biological Research Station), where we explained about our need to find a suitable location for a LOFAR station. He was the one who then gave us a hint to look in the area near the customs station, which is exactly where the construction work is now taking place.


Friday 13 May 2011

KAIRA site panorama

It's Friday, so it's time for a pretty KAIRA/LOFAR picture. Today we have a real treat. Try clicking on this panorama for view of the KAIRA HBA site.

This montage photograph was taken on the 10th May 2011, before we loaded up the raised tile (foreground) with snow. In the background on the left-hand side is Saana and the main ridge in the centre background is Pikku Malla. In the middle distance it is possible to make out the ground tile. Once testing has been completed, both of these tiles will be removed to make way for the main KAIRA system.

More news and images will be posted during the weekend, and there will be lots more next week!

PS: Sorry for the issues with the web log not updating correctly. This is a problem with the Blogger Host that we are using. They are working on the problem and hopefully things will be fixed soon! ( LINK )

Photo: D. McKay-Bukowski

Thursday 12 May 2011

Destroying LOFAR — Load #9

Our vendetta against this LOFAR HBA tile continues. Will the polar caps disappear before we find enough snow to crush this thing? Here the digger drives up with another tonne of snow.

The snow goes on and there is no collapse. The total mass on the tile now exceeds seven-and-a-half tonnes.
Load   Mass    Total
#1 400kg 400kg
#2 550kg 950kg
#3 950kg 1900kg
#4 1050kg 2950kg
#5 850kg 3800kg
#6 1400kg 5200kg
#7 500kg 5700kg
#8 1050kg 6750kg
-100kg 6650kg
#9 1000kg 7650kg

But wait! Just after the load goes on, there is an ominous 'CRACK'. It was not clear if that was part of the HBA failing, or whether it was the settling of the packed snow. We switch off the digger to wait and listen.

After all the noise of the great machine, there is deathly silence. This part of the world can be so still and tranquil. And yet there is a palpable tension in the air.

Like every load, we check the tile from different angles and take photographs. But now there is a hint that something might have happened.

However, despite the sound the tile looks intact. There is no partial collapse of the wall and the mound looks intact, as if there was no deformation. But then we notice the underside of the tile.

Around the location of Cell 10 and 11 (an HBA tile comprises 16 cells), there is slight bulge that was not there after the application of Load #8. The displacement seems to only be about 5cm, but this is the first deformation that we have seen since the testing started.

The catastrophic failure hadn't happened. But at least were were now seeing some form of distortion due to the loading. It was also noted that the distortion was occurring in a single place, which was more-or-less directly under the location where the digger was placing the snow. As a consequence, this was also the area where the snow had been piling up the highest so it was no surprise that the flexing was only being seen there.

But then what happened next took us all by surprise.

Tune in tomorrow for the report!

Destroying LOFAR — Load #8

I don't know why I started this web log series with the title 'Destroying LOFAR'. By this stage we're seriously wondering if LOFAR is actually indestructible and if the 'monoliths' from the Arthur C Clarke book '2001 — A Space Oddessy' were actually in fact some sort of LOFAR HBA tile?

Thinking that the last load was a bit light, we've gone back to one-tonne loads of snow from the digger bucket.

The mound grows ever higher.

However, we also had our first avalanche. It wasn't much, but a few chunks of ice fell away from the back of the tile (opposite the side that the digger was approaching from). We did a check and estimated that the mass of lost material was probably somewhere in the range of 50 to 100 kg. Even so, we're still over six-and-a -half tonnes of snow and ice on our HBA tile.

Load   Mass    Total
#1 400kg 400kg
#2 550kg 950kg
#3 950kg 1900kg
#4 1050kg 2950kg
#5 850kg 3800kg
#6 1400kg 5200kg
#7 500kg 5700kg
#8 1050kg 6750kg
-100kg 6650kg

Still need more snow.

Destroying LOFAR — Load #7

Surely we must be getting close? Or would our snow-supply have melted away before we finished the task.

Another load (albeit smaller this time) was added to the pile.

This photograph does a good job of showing the problem we're now facing. We've more-or-less covered the entire tile surface. There are still no signs of it collapsing and we are starting to wonder if we need some alternative testing method.
Load   Mass    Total
#1 400kg 400kg
#2 550kg 950kg
#3 950kg 1900kg
#4 1050kg 2950kg
#5 850kg 3800kg
#6 1400kg 5200kg
#7 500kg 5700kg

Destroying LOFAR — Load #6

Did we say were were getting serious last time? Hmm... that was nothing.

This time we threw on 1400kg of snow taking the total well beyond the five-tonne mark. This was the heaviest load yet applied, and it came down hard on the tile. However, as you can see, the tile held up just fine.

Just for reference, the frameset is 1.0 metres high and the tile itself is about 0.5 metres or so. The summit of our snow mountain was about 1.5 metres high at this point — some 3 metres off the ground. It was certainly impressive to see.

Load   Mass    Total
#1 400kg 400kg
#2 550kg 950kg
#3 950kg 1900kg
#4 1050kg 2950kg
#5 850kg 3800kg
#6 1400kg 5200kg

Destroying LOFAR — Load #5

Okay... this is getting serious. Would we actually be able to damage a tile with snow? Undeterred (so far) we press on with another load.

The load is piled on to the tile

And we are now well over three-and-a-half tonnes with no ill-effect.

In fact, the mountain of snow is now so high, that it is getting difficult to reach the top to dump more snow (we didn't expect this eventuality). As a result, we decided to spread the snow out a little bit, to allow us to put even more loading on it. This was done by getting the digger to push the summit of our HBA-snow-mountain out to cover more entire of the tile surface.

This work was the first time we noticed any deformation. As the digger pushed the snow across the pile, there was a lateral movement of the tile, where the base remained fixed to the frame, but the top of the tile would move laterally as the snow was pushed. This lateral movement was small, but it was distinct and we thought for a moment that the vertical loading capacity of the tile would be fine, but that the tile would concede when subjected to a shearing force.

But no.

The tile remained intact... and we went off to get more snow. The score now stands are:
Load   Mass    Total
#1 400kg 400kg
#2 550kg 950kg
#3 950kg 1900kg
#4 1050kg 2950kg
#5 850kg 3800kg