Re: Detector support and flange

From: Victor Guarino <>
Date: Wed Jul 14 2004 - 16:41:59 CDT

Hi Hans,
Thanks for your comments. On my next trip to FNAL we should discuss these

The design of the support structure is not meant to be the final design. A lot
of work can be done on this. As Ang does the analysis remember that one of the
design criteria that we used was that the support structure would not induce any
additional deflections and stresses in the sphere. The sphere is supported
along its midsection in order to eliminate bending stresses in the sphere but
supporting it on the lower half of the sphere. Also, in the structure that I
showed there are 16 support points in order to distribute the weight and there
by reducing bending stresses in the sphere. A much different structure can be
developed if the sphere acts as a structural member but this requires increases
in the thickness. The support structure also deflected a very small amount,
fractions of a mm if I remember correctly. I think deflections are important
because we don't want to flex our flange joint. If we are going to take apart
this sphere then only the bottom sphere can act as a structural member so you
don't get the full benefit that the spherical geometry provides.

I agree that its possible to fabricate the sphere. I found several companies
that make spherical heads for tanks who could do this work However, it is not a
simple task. The fabrication has to occur on sight and the flange is not
trivial. I believe that dismantling the entire sphere to get at the pmts will
be a difficult and time consuming task. If we need to get at the PMTs often
then we should think about alternative designs. We'll have to buy more mineral
oil but that is probably off-set by savings in fabrication costs and the
maintenance costs associated with dismantling the sphere. A simple tank
provides easy access and maintenance.

Hans Jostlein wrote:

> Hi, Vic,
> thanks for your response.
> I am still interested in a strut type support, and asked Ang Lee to do some
> very basic FEA .
> Presumably I will learn what you already learned, but I'm curious.
> The struts are meant, of course, to take the forces directly to the 4
> support points.
> We'll see.
> On the shape, I am sure you realize that a cylinder will weigh 50% more
> than a sphere of the same diameter and height.
> That's a big tax to pay!
> I also don't quite see why a cylinder would be easier to support.
> Basically, I don't understand the force flow in your support structure.
> Maybe you can show me next time you are here at Fermilab.
> The fabrication should be no issue at all;
> Chicago Bridge and Iron, as you know, makes those vessels in this and
> larger sizes all the time,
> mostly for water towers.
> They made the 40 ft diameter Miniboone sphere.
> I am also attaching a picture with a couple of comments on the large vessel
> flange.
> Maybe you already drew the flange that way.
> Greetings.
> Hans
> ----- Original Message -----
> From: "Victor Guarino" <>
> To: "Hans Jostlein" <>
> Cc: "David Reyna" <>; <>; "Jim
> Grudzinski" <>
> Sent: Friday, July 09, 2004 3:14 PM
> Subject: Re: Detector support concept
> > Hi Hans,
> > Thanks for your input. This is very similar to the structure that we
> > originally examined. We moved to the structure that I showed in our paper
> for
> > several reasons. First, the structure induced stresses into the sphere
> which
> > required it to get thicker. Second, we are planning on moving the
> detector on
> > Hilman rollers located at discrete points. Therefore, the structure has
> to be
> > very stiff in order to transfer the load from any structure to the support
> > points at the Hilman rollers. I didn't want the structure to deflect to
> much
> > between the hilman roller supports because this would induce stresses into
> the
> > outer steel sphere. I did some simple FEA modeling of the structure but
> of
> > course further optimizing can still occur.
> >
> > I have been thinking that we should get away from the outer containment
> vessel
> > being a sphere. I have done a lot of work trying to find companies that
> can
> > fabricate such large sphere and there aren't many. Also, if we have the
> > requirement that we need to dismantle the outer sphere in order to get at
> the
> > PMT's then we need some sort of flange so that the sphere can be taken
> apart.
> > This is another complication on the design and fabrication. I think an
> easier
> > way would be to make a simple structure that is similar to an above ground
> > pool. The two acrylic spheres would be suspended inside the pool and a
> > geodesic structure can be constructed around them to support the PMT.
> This
> > type of structure would be easier to support and fabricate and access to
> the
> > PMT's would be a lot easier.
> > Vic
> >
> >
> >
> > Hans Jostlein wrote:
> >
> > > Hi, Vic,
> > >
> > > for your amusement,
> > > I have attached a strut-type detector support concept.
> > > I am not sure I understand your drawing completely,
> > > and I apologize if I have overlooked an important constraint.
> > >
> > > Sincerely
> > >
> > > Hans
> > >
> >
> ------------------------------------------------------------------------
> > > Name: Detector support concept.JPG
> > > Detector support concept.JPG Type: JPEG Image (image/jpeg)
> > > Encoding: BASE64
> >
> >
> ------------------------------------------------------------------------
> Name: detector flange.JPG
> detector flange.JPG Type: JPEG Image (image/jpeg)
> Encoding: BASE64
Received on Wed Jul 14 16:42:06 2004

This archive was generated by hypermail 2.1.8 : Fri Jul 16 2004 - 03:28:13 CDT