Choose Search Type
Search
 
 

Display results as :
 


Rechercher Advanced Search

Latest topics
» ROKC Lampoon
Today at 9:08 am by greg parker

» The Bold and the Italics
Today at 9:06 am by greg parker

» Shirley Temple is Prayer Man According to Duncan McRae
Today at 4:12 am by steely dan

» The Eighth Naval District
Yesterday at 11:33 pm by Hasan Yusuf

» Send Lawyers Guns & Money Pt2
Yesterday at 8:08 pm by barto

» Send Lawyers Guns & Money Pt1
Yesterday at 11:58 am by barto

» JFK Assassination
Yesterday at 7:15 am by jack ferguson

» Lifton on his "new evidence"
Yesterday at 4:47 am by steely dan

» friends student exchange programs
Wed 07 Dec 2016, 12:01 pm by greg parker

Log in

I forgot my password

Social bookmarking

Social bookmarking Digg  Social bookmarking Delicious  Social bookmarking Reddit  Social bookmarking Stumbleupon  Social bookmarking Slashdot  Social bookmarking Furl  Social bookmarking Yahoo  Social bookmarking Google  Social bookmarking Blinklist  Social bookmarking Blogmarks  Social bookmarking Technorati  

Bookmark and share the address of REOPENKENNEDYCASE on your social bookmarking website

Bookmark and share the address of REOPENKENNEDYCASE on your social bookmarking website

RSS feeds


Yahoo! 
MSN 
AOL 
Netvibes 
Bloglines 


Affiliates
free forum
 



Walker bullet and NAA

Page 1 of 2 1, 2  Next

View previous topic View next topic Go down

Walker bullet and NAA

Post by Colin Crow on Wed 11 Dec 2013, 7:56 am

I posted this at Duncan's forum.....

Vincent Guinn appeared before the HSCA and presented his analysis of the bullet fragments and CE 399 using Neutron Activation Analysis.


His testimony and report can be found here.....


http://www.maryferrell.org/mffweb/archive/viewer/showDoc.do?docId=95&relPageId=495


He also examined a sample from the Walker bullet to determine if it could be identified as consistent with the MC ammo used in the assassination. Obviously this was intended to confirm Oswald's failed attempt on Walker in April 1963.


Guinn's argument in his ability to discern between bullets in the shooting was that there was considerable difference in elements between bullets from the same box but little difference within a bullet. ie inter bullet heterogeneity but intra bullet homogeneity. The two elements that can best demonstrate this phenomena are Antimony (Sb) and Silver (Ag).


In his report he provides a table to prove this by testing bullets obtained from the only 4 batches of bullets produced for the MC. These are designated 6000-6003. He tested 14 bullets to prove his point. See Table below.





The values for Sb and Ag in ppm are shown and it can be seen that only one bullet has a Sb amount below 80 ppm. Ag values are much tighter and all bullets range between 6 and 16 ppm.


A later study by Spiegleman et al analysed 30 bullets from lots 6000 and 6003. Their results appear below.





In their study the ranges for Sb were large from ~40-1800 and Ag ranged between 4 and 11.


We can see that these studies that tested 44 bullets from each of the lots produced by WW for the MC. There were no bullets that had Sb levels below 20 ppm and only one with less than 40 ppm. There were no bullets that contained more than 16ppm Ag. No bullets had more Ag then Sb.


Quinn analysed the Walker bullet and compared it to the unfired chambered round discovered in CE 149, the rifle found on the 6th floor.


The results of his analysis appear below.......





I find this most interesting. Quinn obviously separates this result from the fragment/CE399 table as the quantities are vastly different. Also it destroys his assumption of heterogeneity between bullets. The quantities are vastly different from the bullet and fragments from the car/victims/ stretcher. Ag is significantly higher in both bullets and the Sb is significantly lower. These bullets were the only 2 that have more Ag than Sb.


I have some thoughts what this might mean but will open discussion before I influence the debate.

Colin Crow

Posts : 214
Join date : 2013-08-03

View user profile

Back to top Go down

Re: Walker bullet and NAA

Post by Stan Dane on Wed 11 Dec 2013, 12:06 pm

I wonder what the tin (Sn) content of the samples in these Tables was.
 
I understand from another thread ("FBI Special Agent Robert Frazier does Mathematics") that antimony (Sb) was typically used in bullets except during the WWII years when it was replaced by tin. After the war, antimony was again used.
 
CE 573 and 141 have similar chemical assays pertaining to antimony and silver (Ag).
 
What I see here are CE 573 and CE 141 being low in antimony concentration (ppm = parts per million) and high in silver concentration. If analyzed, I would suspect that the tin concentration in these two samples would be higher than the samples from lots 6000-6003.
 
During WWII, nickel was a critical war material and the US Mint was authorized to use silver in place of nickel in coining nickels with 35% silver from 1942-1945.
 
I think CE 573 and CE 141 were made during WWII. That they have higher levels of silver, support this.
 
I'm still thinking about it all and what it may mean.

Stan Dane

Posts : 2331
Join date : 2013-09-03
Age : 63

View user profile https://prayermanleeharveyoswald.blogspot.com/

Back to top Go down

Re: Walker bullet and NAA

Post by Stan Dane on Wed 11 Dec 2013, 8:11 pm

I've completed a once-through "brisk" read of the report referenced in Colin's post above.
 
[Note: Before I retired, I was head of the chemistry department at a large nuclear power facility in the Southwest, responsible for radiochemical analyses, so I'm familiar with the language, methods, and terminology used by Guinn in this report.]
 
Some quick reactions:
 
Stripping away all of the technical fluff and focusing on the results tables, I don't see how Guinn can say with the high level of confidence he asserts that the bullet fragments (CE 567, 840, 842, 843) match up with CE 399. Looking at the antimony numbers for these exhibits, I see much variation:   
 
 
 
The antimony concentrations of CE 573 and 141 shown in Table III (in Colin's post above) are vastly lower than the values of the samples in Table I. Guinn acknowledges this, but then ties these results to those found in lots 6000, 6001, 6002, and 6003 (also shown in Colin's post).

I don't see the tie here (right now at this late hour anyway). CE 573 and 141 have lower levels of antimony than any of the lots from which they are purported to be made from—even accounting for margin of error.

After a cursory review, I can't substantiate Guinn's claims.

Stan Dane

Posts : 2331
Join date : 2013-09-03
Age : 63

View user profile https://prayermanleeharveyoswald.blogspot.com/

Back to top Go down

Re: Walker bullet and NAA

Post by Colin Crow on Wed 11 Dec 2013, 9:51 pm

Stan, it tan, it seems we have data for 48 bullets from lots 6000-6003. Guinn provides data for 18. 14 from his control run, supposedly 2 from the shooting, the Walker bullet and the unfired bullet removed from CE149.

Subsequently Spiegleman's study provides data on 30 bullets from lots 6000 and 6003. Guinn's work was based on 2 claims. 

1. Each bullet had a unique composition  with respect to Sb and Ag compared to any other bullet in the same box.

2. Each bullets composition was homogenous with respect to these elements.

Both of these claims are false. However these studies have provided a database of 48 bullets composition from which we can make some observations.

1. Of the 46 bullets other then the Walker bullet and the unfired cartridge the lowest Sb is 24 and the next lowest 80 ppm. 
2. Ag ranges are tighter, and invariably at much lower
amounts.

The Walker bullet and unfired cartridge are "twins" when taking into account the error limitations they overlap.

Their Sb and Ag amounts place them in a category by themselves. One possibility is that the bullets fired that day that produced wounds were sourced differently from these two. If Oswald shot at Walker it seems unlikely the bullets used in the assassination were his.

Another alternative is that the Walker bullet is not from Oswald but they used a second specimen from it and called it the unfired bullet to try and tie the two events together. The problem is that now they have bullets that  do not match anything like the 46 others that we have results for.

Colin Crow

Posts : 214
Join date : 2013-08-03

View user profile

Back to top Go down

Re: Walker bullet and NAA

Post by Colin Crow on Sat 14 Dec 2013, 2:30 pm

This shows what I am talking about graphically. We have a database of Western MC bullets that have been tested. All tested bullets were from the 4 lots produced by Western in 1954. Those lots were designated 6000, 6001, 6002 and 6003. Those that were tested in Guinn prior to the HSCA appear in the graph with a white border. He tested 2 from lot 6000, and 4 from the other lots. He also obtained Sb and Ag from CE399, fragments from the headshot, the Walker bullet and the unfired bullet in the MC found on the 6th floor. Spiegleman also tested 10 bullets from lot 6000 and 20 bullets from lot 6003.


I have plotted each bullet's Ag amount in ppm. A second variable is the size of the circle for each bullet. The size of the circle represents the Ag/Sb ratio for each. As a general trend those with relatively low Sb amounts appear as larger circles. 





Hopefully you can see that the Walker bullet does not appear to be representative of any of the 1954 lots made by Western.

Colin Crow

Posts : 214
Join date : 2013-08-03

View user profile

Back to top Go down

Re: Walker bullet and NAA

Post by Stan Dane on Sat 14 Dec 2013, 5:12 pm

My observations:

1) There is a general decreasing trend in silver (Ag) levels across lots 6000 to 6003, excluding outliers.


2) If circle size is inversely proportional to antimony (Sb) concentration, then there is a generally increasing trend (excluding outliers) across lots 6000 to 6003.


3) The "CE99" sample (I presume you meant 399) and the sample annotated "Head" (CE 843, the metal fragments removed from the President's head at the time of the autopsy), both show a close correlation on the graph in terms of Sb and Ag concentrations. But when I observe the actual values from Table I of Guinn's report, I see a wider gap between the two for antimony: CE 399 – Sb 833 (
±9) ppm; CE 843 – Sb 621 (±4) ppm. The Ag values are essentially identical.

4) The Ag concentrations are very high and the Sb concentrations are very low in both the CE 141 and the Walker bullet samples (outside the margin of error of the tests) relative to the entire sample population of lots 6000-6003.


I agree with your conclusion, Colin "that the Walker bullet does not appear to be representative of any of the 1954 lots made by Western," and I would add CE 141 to this as well.

Stan Dane

Posts : 2331
Join date : 2013-09-03
Age : 63

View user profile https://prayermanleeharveyoswald.blogspot.com/

Back to top Go down

Re: Walker bullet and NAA

Post by Colin Crow on Sat 14 Dec 2013, 8:02 pm

Yes Stan,
Those two bullets are almost as good a match as any that were tested. And yet so different from anything else. Still trying to get my head around the possibilities.

Colin Crow

Posts : 214
Join date : 2013-08-03

View user profile

Back to top Go down

Re: Walker bullet and NAA

Post by Colin Crow on Tue 17 Dec 2013, 5:08 pm

Update.

Here are the means and SD for the dataset across all lots.

Sb Mean 384.1 ppm SD 444.98      
Ag Mean 8.12 ppm   SD 2.60
Ag/Sn Mean 0.06  SD 0.06


Walker Bullet Sb 17 ppm   Ag 20.6 ppm   Ag/Sn 1.21
Unfired Round Sb  15 ppm   Ag 22.4 ppm   Ag/SN 1.49

The Ag value alone in these bullets is between 3 and 4 SD from the Western dataset. About 0.1% chance that it is matched to the dataset.

We can use Sb as a variable as there was only a weak association between the two elements across the sample set (R2=0.0236). If we use the Ag/Sb as the variable the Walker bullet (and unfired cartridge) is more than 20 times the SD from the Western dataset.


I have emailed Ken Rahn, staunch defender of Guinn, to ask for his comment.

Colin Crow

Posts : 214
Join date : 2013-08-03

View user profile

Back to top Go down

Re: Walker bullet and NAA

Post by Stan Dane on Wed 18 Dec 2013, 8:31 am

There's something that bothers me about the antimony (Sb) levels in Western data set (lots 6000-6003). When the standard deviation (a measure of the spread of the data) is larger than the mean of the sample population, it suggests that the data are not distributed normally. In this case, ±1 standard deviation (one sigma) from the mean—where 68% of the data falls if the distribution is normal—would have -61 ppm and 829 ppm as the one sigma boundary limits. Since antimony levels cannot be less than zero, this data is not distributed normally. Other statistical approaches have to be used.
 
My intent here is not to veer off into statistical never-never land, but Sb values are all over the map, too much so to be overly precise in forming any conclusions using normal statistics. But why is this? I got to believe that when these lots were made, Sb was alloyed into the lead in certain, well-defined percentages. It's easy to take for granted that alloys are uniformly mixed, but could it be that Sb is not well mixed throughout the lead metal itself?

Colin Crow wrote:Guinn's argument in his ability to discern between bullets in the shooting was that there was considerable difference in elements between bullets from the same box but little difference within a bullet. ie inter bullet heterogeneity but intra bullet homogeneity. The two elements that can best demonstrate this phenomena are Antimony (Sb) and Silver (Ag).
Perhaps Guinn's assumption of "intra bullet homogeneity" is invalid for antimony in a lead matrix?
 
I did some checking and I found this discussion on lead-antimony in the metallurgy of bullets (emphasis mine):
At typical lead-pot temperatures (ca. 700 F), antimony is only moderately soluble in lead alloys, and as the temperature drops, the solubility of antimony is markedly lower than that of tin. At the eutectic temperature for a binary lead-antimony alloy (484 F), only 3.5% antimony is soluble (note that this is 123 F hotter than of the tin eutectic temperature, but the antimony solubility is less than 1/5 that of tin). At room temperature the equilibrium solubility of antimony in lead is only 0.44%. The precipitated antimony appears as small rods, at the grain boundaries and within the grains themselves. Electron micrographs of lead-antimony alloys clearly show discrete particles of antimony surrounded by a matrix of lead-rich solid solution.
...
It is important to recognize the antimony hardens lead alloys by a fundamentally different mechanism than does tin. Antimony hardens the alloy by precipitation of a separate crystalline antimony phase, which reinforces the squishy plastic lead phase that’s in between the hard antimony crystals. These alloys tend to be brittle because the plastic (squishy) lead phase gets its hardness from the reinforcing hard antimony rods. As the matrix gets deformed the brittle antimony rods shear off and the soft metal fails. In the case of the lead-tin alloys, the tin is more uniformly distributed throughout the matrix, making the matrix itself harder, so plastic deformation of the alloy is more uniform and progressive, not the slip/shear of lead-antimony alloys.
 
Reference: "From Ingot to Target: A Cast Bullet Guide for Handgunners"
http://www.lasc.us/Fryxell_Book_Chapter_3_alloySelectionMetallurgy.htm
If the antimony is indeed not uniformly mixed within the lead but exists in discrete areas throughout the bullet, then the sample location from the bullet (assuming the entire bullet is not sampled) could be critical. In other words, if the sample comes from an antimony-rich portion of the bullet (a small rod), then the results could be far greater than if the sample was taken from an antimony-depleted area.
 
Just some thoughts here. I may look at this later and decide it's crap.
 
I still agree with Colin that the Walker Bullet and the Unfired Round are off into an area by themselves, but I base this mainly looking at the Ag concentrations. I just don't trust what the Sb results are telling us at this point.
 
My big takeaway, however, is this:
 
The HSCA trots out some expert who claims that the bullets/fragments that hit Kennedy came from "Oswald's rifle" (I couldn't substantiate his claim using his own data, but who the hell am I). Then this Texas A&M University researcher, Spiegelman, comes long and says that the fragments could have come from more than two bullets. Hmm....

I'm very skeptical of any witness for the government. It's not that I don't trust them...well, I don't trust them.

Stan Dane

Posts : 2331
Join date : 2013-09-03
Age : 63

View user profile https://prayermanleeharveyoswald.blogspot.com/

Back to top Go down

Re: Walker bullet and NAA

Post by Colin Crow on Wed 18 Dec 2013, 11:19 pm

Stan, let me preface this by saying stats is not my strongest suit. I do get what you are saying about Sb not being normally distributed. For that reason I am using Ag as the primary variable of interest. It is perhaps this element that is the best indicator to focus on. As I said there is only a weak positive correlation between the two elements in the Western MC bullets tested. For this reason I also used the Ag/Sb as a secondary indicator. Especially as the Sb ppm for both bullets of interest are the lowest detected.

Apparently Sb is used in bullet leads at higher concentrations to harden bullets. Regarding the production of lead for jacketed bullets, in the reading I have done, it appears that a "purer" lead is required as the lead core is pushed into the jacket during manufacture. A lower melting point is desirable and therefore the impurities are at much lower concentrations. However sources of lead recovered from car batteries etc that contain Sb can be blended into the  mix as long as the "lead" does not contain an upper threshold of impurities. This is why the jacketed bullets can contain ppm values for Sb.

Obviously there is heterogeneity in the mix as we can see in the Western 6000-6003 data set. Also it seems due to the different melting point temperatures between elements a "cellular" structure is formed as the "lead" cools. Impurities such as Sb appearing around the outside and "grains" can be formed. These can be relatively large in size. The smaller the metal piece assayed by NAA the greater the error rate. I think for the larger pieces analysed (50mg) this is not so much of a problem as these might contain 3 discreet grains. Smaller pieces, such as the Connally wrist fragment are more problematic however.

It is interesting to note that early on the FBI spectrographic analysis indicated the Walker bullet had a different lead composition. Even given the limitations of this method it is consistent with what we see from NAA.


Colin Crow

Posts : 214
Join date : 2013-08-03

View user profile

Back to top Go down

Re: Walker bullet and NAA

Post by Stan Dane on Thu 19 Dec 2013, 5:52 am

Colin Crow wrote:It is interesting to note that early on the FBI spectrographic analysis indicated the Walker bullet had a different lead composition. Even given the limitations of this method it is consistent with what we see from NAA.

It is indeed interesting. I think it's safe to conclude that the Walker bullet lead is different from the Kennedy/Connally bullet lead. Given this, what does it mean?

PS: You might find this article interesting:

"Skepticism Surrounding JFK Assassination Bullet Evidence Focus of Texas A&M Statistician's 50th Anniversary Talk."  http://www.science.tamu.edu/articles/1138/

Note Spiegelman's opinion that Guinn's "lead-don't-lie-so-Oswald-done-it" conclusions were "overstated and not as overwhelmingly certain as it was presented."

Also note this in the article:
It was the Kennedy assassination that popularized the use of a method of forensic testing called Comparative Bullet Lead Analysis (CBLA), a technique that uses chemistry to link bullets found at crime scenes to the guns from which they were discharged.

Spiegelman was a leading critic of the method. After years of opposition from him and others, the FBI announced in 2005 that it would discontinue use, stating in a news release that "neither scientists nor bullet manufacturers are able to definitively attest to the significance of an association made between bullets in the course of a bullet lead examination."

Stan Dane

Posts : 2331
Join date : 2013-09-03
Age : 63

View user profile https://prayermanleeharveyoswald.blogspot.com/

Back to top Go down

Re: Walker bullet and NAA

Post by greg parker on Thu 19 Dec 2013, 6:53 am

thanks to all involved in this thread. One of the most outstanding discussions on the subject I've ever come across.

_________________
Mixing Pop and Politics he asks me what the use is
I offer him embarrassment and my usual excuses
While looking down the corridor
Out to where the van is waiting
I'm looking for the Great Leap Forward

            Billy Bragg
-----------------------------
 Australians don't mind criminals: It's successful bullshit artists we despise. 
             Lachie Hulme            
-----------------------------
The Cold War ran on bullshit.
              Me

greg parker
Admin

Posts : 3450
Join date : 2009-08-21
Age : 58
Location : Orange, NSW, Australia

View user profile http:// http://www.amazon.com/gp/product/B00IXOA5ZK/ref=s9_simh_

Back to top Go down

Re: Walker bullet and NAA

Post by Colin Crow on Thu 19 Dec 2013, 7:00 am

Stan, I have read the articles by Rahn and Sturdivan supportive of Guinn and those by Grant and later Spiegleman that were critical. There was also a report on Lead bullet analysis (284 pages) that I read. I emailed, Spiegleman who is a statistician, a couple of days ago and got a brief reply almost immediately. He said the Bullets should be revisited today and analysed again. I got the impression he may not have grasped what I am on about. 

As for what this means, it might be significant.

Colin Crow

Posts : 214
Join date : 2013-08-03

View user profile

Back to top Go down

Re: Walker bullet and NAA

Post by Colin Crow on Thu 19 Dec 2013, 7:09 am

Thanks Greg. There is more. I just haven't had time to compile a descent post. There are other interesting documents I can post. I wonder if the original spectrographic analysis results are available. It seems the FBI deep sixes them unfortunately. 

It raises all sorts of possibilities. From a substituted CE 399 to the use of older ammo that might result in a "short" round if older ammo was used. It also destroys the myth that Oswald only bought one box of ammo and that he only had 4 bullets left IMO.

Colin Crow

Posts : 214
Join date : 2013-08-03

View user profile

Back to top Go down

Re: Walker bullet and NAA

Post by Stan Dane on Thu 19 Dec 2013, 7:36 am

Thanks as well, Greg. Colin deserves a ton of credit for probing into this. Great research.

Stan Dane

Posts : 2331
Join date : 2013-09-03
Age : 63

View user profile https://prayermanleeharveyoswald.blogspot.com/

Back to top Go down

Re: Walker bullet and NAA

Post by Guest on Fri 20 Dec 2013, 2:53 pm

"It was the Kennedy assassination that popularized the use of a method of forensic testing called Comparative Bullet Lead Analysis (CBLA), a technique that uses chemistry to link bullets found at crime scenes to the guns from which they were discharged.

Spiegelman was a leading critic of the method. After years of opposition from him and others, the FBI announced in 2005 that it would discontinue use, stating in a news release that "neither scientists nor bullet manufacturers are able to definitively attest to the significance of an association made between bullets in the course of a bullet lead examination."

This article is a bit misleading, as CBLA is used to link crime scene bullets to same calibre and manufacturer bullets found in a suspect's possession, often in the absence of a firearm.

While the merits of CBLA can be questioned in the case of mass produced bullets, the JFK case was an ideal case to apply CBLA to. Only four million rounds of 6.5mm Carcano ammunition were manufactured by the Western Cartridge Co. in four lots, of which, only two lots were available for sale to the public.

I still believe the original Walker bullet, described by DPD detectives as "steel jacketed", and, according to Gen. Walker, not the bullet placed in evidence by the WC, was likely a cupro-nickel jacketed 6.5mm Carcano bullet manufactured in Italy by SMI and issued to Italian troops. These had found their way into the surplus market after WWII and would likely have been available in the USA. This is, of course, if we are to believe Gen. Walker was actually fired upon by a 6.5mm Carcano.

I have always found the claim of WC defenders that DPD detectives were simply "mistaken" in their observation of the Walker bullet being "steel jacketed" to be quite difficult to swallow, and FBI SA Robert Frazier's interpretation of their "mistaken" observation even more difficult to swallow.

Outside of the difficulty of communicating test results to juries and prosecuting attorneys, did the FBI ever give a truly definitive reason why CBLA was abandoned in 2005, after decades of use?

Guest
Guest


Back to top Go down

Re: Walker bullet and NAA

Post by Stan Dane on Fri 20 Dec 2013, 4:17 pm

Traveller11 wrote:"It was the Kennedy assassination that popularized the use of a method of forensic testing called Comparative Bullet Lead Analysis (CBLA), a technique that uses chemistry to link bullets found at crime scenes to the guns from which they were discharged.

Spiegelman was a leading critic of the method. After years of opposition from him and others, the FBI announced in 2005 that it would discontinue use, stating in a news release that "neither scientists nor bullet manufacturers are able to definitively attest to the significance of an association made between bullets in the course of a bullet lead examination."

This article is a bit misleading, as CBLA is used to link crime scene bullets to same calibre and manufacturer bullets found in a suspect's possession, often in the absence of a firearm.

While the merits of CBLA can be questioned in the case of mass produced bullets, the JFK case was an ideal case to apply CBLA to. Only four million rounds of 6.5mm Carcano ammunition were manufactured by the Western Cartridge Co. in four lots, of which, only two lots were available for sale to the public.

I still believe the original Walker bullet, described by DPD detectives as "steel jacketed", and, according to Gen. Walker, not the bullet placed in evidence by the WC, was likely a cupro-nickel jacketed 6.5mm Carcano bullet manufactured in Italy by SMI and issued to Italian troops. These had found their way into the surplus market after WWII and would likely have been available in the USA. This is, of course, if we are to believe Gen. Walker was actually fired upon by a 6.5mm Carcano.

I have always found the claim of WC defenders that DPD detectives were simply "mistaken" in their observation of the Walker bullet being "steel jacketed" to be quite difficult to swallow, and FBI SA Robert Frazier's interpretation of their "mistaken" observation even more difficult to swallow.

Outside of the difficulty of communicating test results to juries and prosecuting attorneys, did the FBI ever give a truly definitive reason why CBLA was abandoned in 2005, after decades of use?
I don't think the referenced article is misleading at all.
 
Comparative bullet-lead analysis (CBLA) is not about determining the caliber of bullets or other physical determinations such as rifling marks, etc. That continues to be a vital part of the forensic processes. CBLA is about making chemistry determinations of the bullet lead itself. This is what has been discredited and abandoned.
 
Here's what the FBI had to say about the matter in their September 1, 2005 press release: "FBI Laboratory Announces Discontinuation of Bullet Lead Examinations":

http://www.fbi.gov/news/pressrel/press-releases/fbi-laboratory-announces-discontinuation-of-bullet-lead-examinations


"In 2002, the FBI asked the National Research Council (NRC) of the National Academy of Science to have an independent committee of experts evaluate the scientific basis of comparative bullet lead analysis.
...
"The NRC found that the FBI Laboratory's analytical instrumentation is appropriate and the best available technology with respect to precision and accuracy for the elements analyzed. It also found that the elements selected by the FBI for this analysis are appropriate. The NRC expressed concerns, however, relating to the interpretation of the results of bullet lead examinations."

 
Ah, it seems the devil is in the details, or in this case, the interpretation of the details.
 
Chemical & Engineering News published this article on March 2, 2009: "Board's Report Leads To Overturn Of Murder Conviction":

https://pubs.acs.org/cen/government/87/8709gov1a.html

 
The gist of the article is that a man in Florida was convicted of murder, and after serving 10 years, he was released because of a report issued by the National Academies' Board on Chemical Sciences & Technology (BCST) which dismissed a long-standing protocol for bullet-lead analysis used by the FBI. They go on to say that while the chemical analyses, instrumentation, and analytical techniques were appropriate and reliable,

"...the report throws into question the statistical tests that FBI forensic scientists have been using to make sense of the results of elemental analyses from different bullet samples. In addition, the report suggests that the significance of the established relationship between bullet-lead samples based on the results of the chemical and statistical analyses could be easily and inadvertently misrepresented by prosecutors or misinterpreted by a trial jury. Because this can threaten the fairness of a trial, the report recommends that CBLA testimony be used with caution."

 
That's exactly what we've been trying to do here in this thread: interpret results and make sense of the statistics. And I, for one, have been having a tough time of it.

The real bottom line though: the attempts used by the Warren Commission, the HSCA, or any other arm of the government to prove that the bullets they found are linked to Oswald are bogus.

Stan Dane

Posts : 2331
Join date : 2013-09-03
Age : 63

View user profile https://prayermanleeharveyoswald.blogspot.com/

Back to top Go down

Re: Walker bullet and NAA

Post by Albert Rossi on Fri 20 Dec 2013, 5:30 pm

I am neither a firearms expert nor a metallurgical chemist, but it is my understanding that the whole statistical basis for drawing any kind of conclusions in this area has been undermined most forcefully by an article which appeared in Journal of Forensic Science, July 2006, by two scientists at the Lawrence Livermore laboratory, Erik Randich and Patrick Grant.   If I have understood the results of that study correctly, statistical correlations either in the direction of heterogeneity (different lots are marked by distinct compositional percentages of elements) or homogeneity (a given lot will have a fairly uniform compositional breakdown) cannot be shown to hold within any reasonable sigma (standard deviation) such that something like a fingerprint could be developed.  Further, they showed that within a single bullet, the percentages of trace elements could vary widely depending on the section or area from which it was taken.  And finally, they demonstrated that Guinn's special pleading (that MC bullets were different from other rifle bullets in the amount of heterogeneity they exhibited) was fallacious.   The problem, of course, was not the accuracy of the method of measurement but rather its application, which made several erroneous assumptions about the chemical processes governing the production of bullet lead.

It seems to me that later work has tended to reinforce this article's conclusions, despite the tenuous efforts of Rahn and Sturdivan to keep Guinn's findings alive.

Albert Rossi

Posts : 417
Join date : 2013-08-29
Age : 61
Location : Naperville, IL USA

View user profile

Back to top Go down

Re: Walker bullet and NAA

Post by Stan Dane on Fri 20 Dec 2013, 7:18 pm

Albert Rossi wrote:I am neither a firearms expert nor a metallurgical chemist, but it is my understanding that the whole statistical basis for drawing any kind of conclusions in this area has been undermined most forcefully by an article which appeared in Journal of Forensic Science, July 2006, by two scientists at the Lawrence Livermore laboratory, Erik Randich and Patrick Grant.   If I have understood the results of that study correctly, statistical correlations either in the direction of heterogeneity (different lots are marked by distinct compositional percentages of elements) or homogeneity (a given lot will have a fairly uniform compositional breakdown) cannot be shown to hold within any reasonable sigma (standard deviation) such that something like a fingerprint could be developed.  Further, they showed that within a single bullet, the percentages of trace elements could vary widely depending on the section or area from which it was taken.  And finally, they demonstrated that Guinn's special pleading (that MC bullets were different from other rifle bullets in the amount of heterogeneity they exhibited) was fallacious.   The problem, of course, was not the accuracy of the method of measurement but rather its application, which made several erroneous assumptions about the chemical processes governing the production of bullet lead.

It seems to me that later work has tended to reinforce this article's conclusions, despite the tenuous efforts of Rahn and Sturdivan to keep Guinn's findings alive.
Thanks for the tip, Albert. The report you refer to is "Proper Assessment of the JFK Assassination Bullet Lead Evidence from Metallurgical and Statistical Perspectives," by Erik Randich and Patrick Grant, July 2006.
 
I went and "rented" the 11-page report for 48 hours (much cheaper than buying it) so I'll give it a good going over and make a post on my observations soon. Colin probably already has this.

Stan Dane

Posts : 2331
Join date : 2013-09-03
Age : 63

View user profile https://prayermanleeharveyoswald.blogspot.com/

Back to top Go down

Re: Walker bullet and NAA

Post by Colin Crow on Fri 20 Dec 2013, 8:09 pm

I need a little time to respond to your concerns. I think at this point the best thing is to be aware that there is nothing wrong with the NAA results, it's in their interpretation and underlying assumptions.

By the way, if any of you need scientific articles that require payment, let me know. Many I can get free and send to you.

Colin Crow

Posts : 214
Join date : 2013-08-03

View user profile

Back to top Go down

Re: Walker bullet and NAA

Post by Colin Crow on Fri 20 Dec 2013, 8:19 pm

I have found the FBI spectrographic analysis via Rahn's website. I am not convinced it hasn't been "massaged".

Also here is another clue. I was talking to a colleague a few days ago. A fellow scientist about what might have happened with Guinn's analysis when the Walker bullet did not match. I wondered if they called in the unfired round after the original tests in order to link it to Oswald. I later found this......bingo. The unfired round, CE141 was prepared the following day.


Colin Crow

Posts : 214
Join date : 2013-08-03

View user profile

Back to top Go down

Re: Walker bullet and NAA

Post by Stan Dane on Sat 21 Dec 2013, 3:34 am

My synopsis of "Proper Assessment of the JFK Assassination Bullet Lead Evidence from Metallurgical and Statistical Perspectives," by Erik Randich and Patrick Grant, July 2006.
 
The report opens up with a short review of what is known concerning the JFK assassination ballistics, the official conclusions and some open issues.
 
Because of "extremely large variability" of antimony content in Mannlicher-Carcano (MC) full-metal-jacketed (FMJ) bullets manufactured by Western Cartridge Company (WCC), Vincent Guinn told the HSCA in 1978 that he was able to individualize bullets, not from lot to lot, but definitely among bullets from the same box. Consequently, he testified that there was highly reliable evidence for two—and only two—bullets among the various fragments of lead recovered from the assassination, that both bullets were from MC rounds, and further that it was very probable that the three metal fragments recovered from John Connally's wrist (CE 842) derived from CE 399. Likewise, the two lead fragments from JFK's brain (CE 843) were from another single bullet, but definitely not CE 399.
 
This assessment by a distinguished scientist gave much support for and reinforcement of the single-bullet theory. Guinn, however, did not interpret his measurements with knowledge of basic metallurgy of lead alloys and bullet manufacturing practices, so his conclusions are unsupported from those perspectives. He also disregarded his own previous findings where he quantified antimony, copper, and arsenic in 75 lots of bullets from various manufacturers. He found less than half the lots were uniquely characterized by the concentrations of these elements, considered individual or together. Finally, he didn't realize that the constraints of his small sample size contributed to variability in his measurements because some elements he quantified were not distributed uniformly within the samples at this scale.
 
It's wasn't clear to the authors, therefore, how Guinn could make such definitive statements about bullet and fragments from the quantification of only one element, antimony. In addition, had more appropriate statistical analyses of those data been performed, his conclusion of two bullets would have been further weakened as well.
 
Several pages of the report are devoted to an in-depth discussion of data collection to identify and quantify major, minor and trace elements, interpretation of the data to determine the likelihood of a common origin, and the statistical analyses and methods to support the interpretation process. I didn't delve too deeply into this section, but the report points out that data collection and data interpretation, while they sometimes overlap, are different areas of expertise, requiring a different knowledge base, and hence different qualifications for the person performing the task (I know this to be true from my experience). Guinn was highly qualified to do the former, but perhaps not so well qualified to do the latter.
 
The next four-five pages focus on bullet and bullet lead production, and metallurgy. There are three basic types of lead materials used for bullet production: (1) lead-tin alloys are used for individually cast bullets; (2) lead-antimony alloys (which represent the vast majority of commercial bullets) are used for extruded or swaged bullets; and (3) soft (not intentionally alloyed) lead which is used for jacketed bullets.
 
The lead used for the cores of jacketed MC bullets was specified to be #0 soft lead (98.85% pure). WCC MC bullet jackets were made with CDA 220 bronze which is 90% copper and 10% zinc. The jacket is produced as an empty shell and the lead core is physically pressed into it. Although either lead-tin or lead-antimony can be used for the core, manufactures typically don't incur the extra cost because the core is merely a high-density filler material. Today, the normal specifications for soft lead require <0.1% antimony. (Silver is an impurity that is difficult for some smelters to remove except through dilution.)
 
Guinn noted that the MC ammunition "was found to differ sharply from typical bullet leads" because "although individual bullets were found to be fairly homogenous in their antimony and silver concentrations, they differed greatly from bullet to bullet amongst samples taken from the same box." Thus, "it would be possible to distinguish one bullet (or bullet fragment) from another, even though they both came from the same box of MC cartridges."  
 
The report goes on to show that the antimony, silver, and copper levels are quite similar and not sharply different, remarking that Guinn was noting large relative differences (but small absolute-value differences) in small quantities of the measured antimony and copper. The common occurrence of microsegregation explains Guinn's observations and the variability in his data (covered below).
 
The JFK FMJ MC bullets contained a soft lead core. The only exposed lead is on the tail, or back end, of the bullet. Guinn's samples of the assassination evidence weighed from 5.4 to 50.5 mg. The samples of the four WCC lots for comparison purposes weighed from 44-58 mg. All of the drillings were taken from the tails of the bullets.
 
The question that must be answered in any review of Guinn's work is whether samples of these sizes were truly representative of the average bulk composition of the lead core. For example, if the bulk material was not homogenous on the 50 mg scale (and much smaller down in the JFK fragments), then variation among the samples from the same bullet should be expected. Likewise, comparison of samples from different bullets with only slightly differing compositions might show overlap. This could lead to erroneous conclusions when assessing common sources of samples being compared.
 
The report then discusses the small grains or crystals of the lead. Basically, alloyed materials, such as antimony, generally are highest in concentration at the grain boundaries (microsegregation) because they are more open in structure and can accommodate non-lead atoms more easily than in the interior of the grains. To visualize this, here are some images of the grains from the WCC MC lots:
 

 
As you can see grain size is highly variable in each sample.
 
Conclusions
 
Basic metallurgical properties of segregations, grain size, and grain-boundary precipitation must be considered when interpreting the data obtained by Vincent Guinn. The authors assert that, from perspectives of standard metallurgical practice and statistical assessment of the NAA measurements (and despite the opinion of Rahn and Sturdivian that their assessment is definitive and puts the matter to rest), a conclusion for only two bullets in the JFK assassination specimens has no forensic basis. The recovered bullet fragments could be reflective of anywhere between two and five different rounds fired in Dealey Plaza that day. Furthermore, the fragments need not necessarily have originated from MC ammunition; the antimony compositions are consistent with any number of jacketed ammunition containing unhardened lead.

Stan Dane

Posts : 2331
Join date : 2013-09-03
Age : 63

View user profile https://prayermanleeharveyoswald.blogspot.com/

Back to top Go down

Re: Walker bullet and NAA

Post by Colin Crow on Sat 21 Dec 2013, 10:21 am

Excellent summary Stan. If I can introduce a simple analogy to explain the difference between these criticisms of Guinn's claims and my position. 

If the bullets were people. Guinn was saying that the fragments recovered from the car, Connally and CE399 were parts of 2 different people. The claimed it was rare to find similar bullets (twins) in WCC MC ammo. He also claimed that the variation within bullets (part of a person) was small (cf DNA). 

Clearly the issues with his work were caused by small sample size of the population (only 14 bullets) and size of the bullet sample (40-50) mg. He relied on Antimony (Sb) as the test variable. His ignorance of grain structure and properties of the antimony during the production process meant his assumptions were somewhat flawed. Spiegleman added to the population data by sampling another 30 bullets. They showed that the expanded data set revealed many bullets that matched. The frequency of "twins" in the population was not rare.

I have produced this graphic to demonstrate this. The bullets that've been sorted according to antimony high to low. Apparent "twins" are designated by a red line over the top. The bullets are labelled according to the following. "G" and "S" represents a Guinn or Spieleman bullet. The next number is lot number and the final number is bullet number.



In my analogy I am suggesting that the Walker bullet and the unfired round are not part of the WCC MC population. Their composition suggests they are not ”human" in the analogy. Furthermore their composition suggests they are "twins".

In summary, Guinn assumed that it was rare for bullets (humans) to have twins and that bullets have DNA that is present throughout.

I am saying that the Walker bullet and the unfired round are not human but are twins.

Colin Crow

Posts : 214
Join date : 2013-08-03

View user profile

Back to top Go down

Re: Walker bullet and NAA

Post by Colin Crow on Sat 21 Dec 2013, 10:56 am

Some points about the lead that is used for production of jacketed bullets. For these bullets the lead is unhardened and "pure". Actually is is about 99.85% Pb so a small amount of impurity is tolerated. My understanding is that the jacket is used to prevent breakup of the lead as it passes down the barrel. 

Lead is produced by smelting of the mineral galena. This contains a reasonable amount of silver.

http://geology.com/minerals/galena.shtml

It is obviously economically advantageous to remove as much silver from the mineral during the smelting process as possible. However a small amount of silver remains in the lead produced. I think this is the source of silver in the soft lead.

There are traces of antimony in galena and so a small amount of this element is likely after the smelting process. In the manufacture of the bullets by WCC the pure lead is augmented by the addition of lead that is recycled from car batteries. They contain 94% lead but 6% antimony. As the lead for bullets can contain 0.5% impurities this allows reduction of the production cost. I believe this is the major source of the antimony content in the bullets. So we have a mixture of the smelted lead and the addition of car batteries in the "melt". Mixing is sufficient to keep the impurity level for each bullet low but does not mix sufficiently to provide for uniformity of antimony in the batch. So pockets of higher antimony content exist in the batch.

As we have seen there is also the formation of boundaries where antimony can concentrate during the cooling process. This is due to the different chemical properties between lead and the other metals present.

I believe this makes silver a better indicator for the source of the bullets. However it also allows use of antimony content as well to be used to differentiate between samples to some degree.

The "pure lead" from smelting alone has antimony less than 20ppm. This would represent lead that has not been augmented by addition of car batteries.

The levels of Sb in the Walker bullet and the unfired round are 17 and 15 ppm respectively. To me this indicates they were produced from a batch that was not augmented. All 46 bullets tested from lots 6000-6003 show evidence of additional antimony added.

It is for these reasons I believe it is likely that those 2 bullet leads were not produced the same way as those used in the assassination.


Last edited by Colin Crow on Sat 21 Dec 2013, 10:20 pm; edited 1 time in total

Colin Crow

Posts : 214
Join date : 2013-08-03

View user profile

Back to top Go down

Re: Walker bullet and NAA

Post by Colin Crow on Sat 21 Dec 2013, 11:38 am

Some graphics to show how the addition of used batteries affects the batch when mixing is not complete.

After smelting some residual silver and antimony in the batch. Both at low levels and relatively homogenous.




When WCC manufactured the bullets they added recycled lead to the pure ingots. This was allowable as long as the lead was 99.85%. Antimony content was allowable up to a value of 0.1% (1000ppm). The increased antimony from batteries raises the Sb level and mixing is not complete. Produces an heterogeneous mix for Sb but still relatively uniform silver. One impact might be a slight lowering of the sliver in ppm by this process as the additional lead from recycled batteries is silver free.



Last edited by Colin Crow on Sat 21 Dec 2013, 8:42 pm; edited 1 time in total

Colin Crow

Posts : 214
Join date : 2013-08-03

View user profile

Back to top Go down

Re: Walker bullet and NAA

Post by Sponsored content Today at 12:57 pm


Sponsored content


Back to top Go down

Page 1 of 2 1, 2  Next

View previous topic View next topic Back to top

- Similar topics

 
Permissions in this forum:
You cannot reply to topics in this forum