New Shield

How to build the perfect shield for your measurements? Discuss it here!
wraith_oz
Posts: 8
Joined: 02 May 2015, 20:15
Contact:

New Shield

Post by wraith_oz » 04 May 2015, 19:25

Having finally got some time to devote to gamma spectroscopy and collected most of the hardware required, the design and construction of a shield was considered a priority. Some time was spent searching for information and pondering what it all meant as well as finding out who sold pure lead in my part of the world and whether they would sell to non-commercial users.

The lead portion of a shield stops most of the incident background radiation but in doing so generates fluorescent x-rays and hence a 'secondary' absorber is needed to remove these. But this secondary absorber also generates its own fluorescent x-rays hence the need for a graded shield where each of the inner shields absorbs the radiation generated by the preceding material.

For a graded shield, cadmium is the preferred material to place inside the lead part to absorb fluorescent lead x-rays but I haven't been able to locate a supplier of cadmium in sheet form (at least at a price I could live with). Looking at tables of x-ray attenuation coefficients it can be seen that tin is the next best material after cadmium but this doesn't seem to be readily available in sheet form either but pewter is, and lead-free pewter is predominantly tin, therefore the inner shield is made of pewter. To absorb the tin fluorescent x-rays copper is used.

The attached files describe the design, construction and characterisation of the new shield as well as calculation of the thickness’s of the inner shield materials.
Attachments
Xray_Mass_Attenuation_Calcs.ods
Calculation of shield thickness
(76.62 KiB) Downloaded 362 times
new_shield_1.pdf
Description of shield
(2.93 MiB) Downloaded 671 times
John Bell from Central Victoria (Australia)

Setup: GS2812 detector/GS-1100 Pro interface; E-MU204 adc; lead (30mm) shield/ 2.0mm lead-free pewter/0.51mm Cu liners

User avatar
Steven Sesselmann
Posts: 693
Joined: 27 Apr 2015, 11:40
Location: Sydney
Contact:

Re: New Shield

Post by Steven Sesselmann » 05 May 2015, 07:42

John,

Thanks for posting, and top marks for design, construction and attention to detail. This is one of the best home made shields I have seen, and I can appreciate the work that went into cutting, melting and machining the ingots.

The results speak for themselves your final background count at 0.04 cps is fantastic, with that you should be able to see trace amounts of radiation stand out above the background.

Maybe you can send me a pm and tell me who supplies the pewter.

Steven
Steven Sesselmann | Sydney | Australia | gammaspectacular.com | groundpotential.org | beejewel.com.au |

Boris
Posts: 36
Joined: 27 Apr 2015, 23:33
Contact:

Re: New Shield

Post by Boris » 08 May 2015, 04:03

wow that is impressive work!

I'ts way lower than my 6 cps and I use 70 mm of lead, 2 mm of Cu and 9mm of aluminium...(as part of the construction).
makes me wonder what would happen if I replace the inside with a 10 mm tin cast...

experiment waiting to happen...

you did an excellent job with a nice finish and result...

Kind regards, Boris.

User avatar
brehwens
Posts: 88
Joined: 27 Apr 2015, 03:49
Contact:

Re: New Shield

Post by brehwens » 08 May 2015, 04:51

Hi John!

Really beautiful shielding, amazing job!

If I understand you correct, the total CPS went from around 100 to 3.6, yes?

I know the feeling of pondering how much lead is enough... at some point it becomes a practical issue also, and one has to think about what samples one will measure, and if the sample geometry can be improved to increase the signal-to-noise ratio. For my 3x3" probe, a 3 cm shield brought the BG down from 500 to 50, but this is still quite high when low activity samples are used (with standard geometry).

Will you be able to fit a marinelli-type container in the castle, maybe some homemade beaker?

That pewter sounds interesting...might look into that....

Great job!!
Karl Brehwens
Eskilstuna, Sweden.

Setup: 5 cm lead castle, 2mm copper lining. Gamma Spectacular Pro 2002, Sound Blaster Live! 5.1,
Primary detector: Scionix refurbished 2x2" NaI(Tl) well detector, 7.5% @ 662 keV

wraith_oz
Posts: 8
Joined: 02 May 2015, 20:15
Contact:

Re: New Shield

Post by wraith_oz » 08 May 2015, 09:07

G'day Karl

Yes that is correct, the background rates dropped from about 100 counts/second to 3.6 counts/second for no shield and with shield respectively.

I realised once construction had started that I wouldn't be able to use a Marinelli style container. But thinking about it, to be able to use one the amount of lead I would have needed would have increased to perhaps 200kg. However with the shielding I currently get I may still be able to measure low activity samples using an extended count period.
John Bell from Central Victoria (Australia)

Setup: GS2812 detector/GS-1100 Pro interface; E-MU204 adc; lead (30mm) shield/ 2.0mm lead-free pewter/0.51mm Cu liners

wraith_oz
Posts: 8
Joined: 02 May 2015, 20:15
Contact:

Re: New Shield

Post by wraith_oz » 08 May 2015, 09:15

Boris

The counts/second also depend on the detector one is using. If your detector is larger or more sensitive than mine you wouldn't expect to see the lower count rate. However better shields may help.

I attached a spreadsheet to my original post which enabled the calculation of the performance of different shield materials. That file was in open document format which excel won't read. An excel version is attached to this message.

kind regards

John
Attachments
Xray_Mass_Attenuation_Calcs.xlsx
Shield liner performance calcs
(31.51 KiB) Downloaded 224 times
John Bell from Central Victoria (Australia)

Setup: GS2812 detector/GS-1100 Pro interface; E-MU204 adc; lead (30mm) shield/ 2.0mm lead-free pewter/0.51mm Cu liners

Boris
Posts: 36
Joined: 27 Apr 2015, 23:33
Contact:

Re: New Shield

Post by Boris » 09 May 2015, 06:26

thanks for the sheet, really good work,

My GS-2812 is outside the shield 53,8 cps and inside 5,46 cps according to becq.monitor. average after 120 sec of measurement.
The strange thing is, some time ago I installed two bronze cylinders also 9 mm wall thickness (copper/tin alloy) but this did not help at all.
What i would like to try is to make a cylinder of used tungsten carbide tool bits, the disadvantage is that it will be like 15 euro's / kilo.
would this help much?

Cheers, Boris.

wraith_oz
Posts: 8
Joined: 02 May 2015, 20:15
Contact:

Re: New Shield

Post by wraith_oz » 09 May 2015, 10:48

Alloys won't work particularly well. The reasons for a graded shield (sorry if I'm teaching you how to suck eggs btw) are that external radiation generates fluorescent x-rays in the lead shield. These have an energy of about 85 keV. The cadmium(or tin) layer is needed to block these but in doing so generates its own fluorescent x-rays with an energy of about 28 keV. The copper layer is used to block these. Thus if you have an alloy of copper and tin you can get the situation where there are tin atoms at the surface generating x-rays with nothing to block them.

The gross counts with and without the shields is one set of numbers but the energy of these counts is also important. There is a graph in the paper I wrote that shows the lead x-rays quite plainly (when measured without the tin and copper shields in place but they disappear completely when the tin and copper liners are used. When I look at my raw data there is very little difference in the counts without and with the Sn/Cu liners in place, 3.72 and 3.48 cps respectively. What is important is where these counts occur.

One other point - when trying to determine the difference that your bronze liner made I think a 120 sec count time may be too short. I did mine for a little over 7 hours which may be overkill but I wanted a good data set to plot the response against energy. Even so, the count in individual bins was too low to plot directly and I ended up aggregating the count into 10 keV bins for plotting.

Kind regards

John
John Bell from Central Victoria (Australia)

Setup: GS2812 detector/GS-1100 Pro interface; E-MU204 adc; lead (30mm) shield/ 2.0mm lead-free pewter/0.51mm Cu liners

Boris
Posts: 36
Joined: 27 Apr 2015, 23:33
Contact:

Re: New Shield

Post by Boris » 09 May 2015, 16:55

Well John, I did longer measurements in the past with this shield (this is the 3.0 version) and the lead and peaks are clear.
A tin layer between the lead and copper, would cure the problem. the basic problem is to get tin in amounts that I can cast.
(pictures in the old forum under hardware)
Maybe a trip to the scrapyard, can help me to some good material...

Kindregards, Boris.

Martin4s
Posts: 9
Joined: 19 Mar 2017, 15:19
Location: Central California
Contact:

Re: New Shield

Post by Martin4s » 21 Mar 2017, 19:10

Looking at this build I have to say it is a great machine skill. I ordered 3 6"X12" Lead low Pewters and 3 Copper sheets the same size. It is Shipping at a amazing 13+ Lbs. Or 6+ Kg. I realise that there needs to be a near zero Tolerance. Am thinking to Wrap the sheets in a Circle and lead Shield around them. We all know Circles are efficient. :-)

Post Reply

Who is online

Users browsing this forum: No registered users and 1 guest