Re: [Phys-l] Alpha decay intensity against distance

[Bernard Cleyet <bernardcleyet@redshift.com>]

Duh .. Now why would I assume the counts were SI instead of CPM.   I  
would necessary verify, as I think the value I calc'd multiplied by  
60 is the dead time, i.e. ~ 240 microsec.   However, and this is what  
I learned from Ludwik,  that dead time only applies to counting rates  
within the order of 10k CPM.  The dead time varies w/ the rate.  It  
is very long a very low rates < 100 CPM and shorter than 200  
microsec. at rates above 10k CPM.  Ludwik (this must be before Dr.  
Edmiston's time) was using counts to illustrate statistics -- He  
didn't know which, I replied Poisson and Denker confirmed and added.   
I don't remember why, and don't want to bother being an archivalist,   
I learned the dead time was millisecs from data LK sent me. OTOH why  
would I need the data, if only to verify L's calc., if he was using  
two source trials.  Anywise, I was amazed and duplicated his  
measurement.  Sho nuff very long.  The clue was he had VERY weak  
sources.  So I did a study w/ several (five?) various diameter tubes  
and several counting rates.  I found, as expected, the DT was a +  
function of the diameter and an inverse of the rate.  I meant to do  
the trials again w/ more care, find the functions by regression, and  
publish, well this explains why I've had no career**.

My last attempt was to do a two source determination of DT with an  
extremely high background, e.g. 10k CPM w/ the sources somewhat less  
active.  Because in earlier measurements I had used sources of  
approximately background rates, I had to find the formula w/o the  
usual zero background assumption.  IT that formula would apply.


** Not the only original work I've let languish.

bc thinks why the DT ~ 1/rate is obvious.



On 2009, Sep 21, , at 14:48, Edmiston, Mike wrote:

> As for dead time, the data in the table are counts per minute, so  
> the dead time is being shown in minutes.  The number listed 4e-6  
> minutes.  In seconds this would be 2.4e-4 s, which is indeed, as  
> Bernard suggested, a couple hundred microseconds.  Dead time was  
> measured in the standard way of counting a split source.  In this  
> case a beta source.  I don't recall for sure, but think each half  
> of the source was about 10,000 cpm, so you count each source alone,  
> then together, and notice that 10,000 plus 10,000 only yields  
> perhaps 17,000 (when counted simultaneously), so dead time is  
> calculated from this.  Note, these count rates for dead-time  
> measurement are made-up numbers, but I think they are in the  
> correct ballpark.