Jump to content


Photo

Authenticating Helmets With XRF Spectroscopy


  • Please log in to reply
64 replies to this topic

#26 Bugme

Bugme

    ADMINISTRATOR

  • Administrators
    • Member ID: 1,726
  • 15,260 posts
  • Gender:Male
  • Location:Marsaxlokk, Malta & Wisconsin Shoreline

Posted 06 September 2013 - 02:05 PM

In case it hasn't been mentioned yet, another thing to consider is that is costs several hundred dollars(I believe it was in the $250.00 range) to have this test run on one helmet. Another reason that this can really only be useful on extremely valuable helmets. Imagine a collection with 50 helmets. The collector has questions on at least 10 of them, so to satisfy his desire to sleep better at night concerning his helmets would cost him $2500.00????

BTW, the test run on the helmets by Hurtgenwald kinda sums up part of the problem. The paint between two helmets with original OD finish has a variable in it. Now that's just two helmets with factory original paint. Now, there is the issue of the paint used to do the insignia. It really is science and I don't think we've arrived to the place where cost equals value for this type of test.



#27 Bob Hudson

Bob Hudson

    Forum Co-Founder (Ret)

  • Administrators
    • Member ID: 2
  • 26,682 posts
  • Gender:Male

Posted 06 September 2013 - 02:20 PM

It case it hasn't been mentioned yet, another thing to consider is that is costs several hundred dollars(I believe it was in the $400.00 range)

 

For half that amount I'll fire up Photoshop and make a COA for any helmet.



#28 just-a-good-ole-boy

just-a-good-ole-boy
  • Members
    • Member ID: 2,533
  • 1,715 posts
  • Gender:Male
  • Location:The Deep South

Posted 06 September 2013 - 02:25 PM

I think it can be a useful tool to weed out some fakes yes, but in my opinion its in the same line as a lead paint test kit. You know if it doesn't have lead then it is most likely a fake, but if it does have lead then maybe its real or maybe it it isn't. I think there are way too many variables to to make this test fool proof. When it comes to painted insignia, I just dont see how there could ever be a complete database for known original used paint. I am sure there are many differences between paint that might have come from different counties. British, German, Italian, US paints etc etc. Paints from any of these countries theoretically could end up on an M-1. Not to mention if some local dirt ended up in the paint like on some camo jobs...I am sure that gives a different reading as well.

 

Steve


Edited by just-a-good-ole-boy, 06 September 2013 - 02:36 PM.


#29 Hürtgenwald

Hürtgenwald
  • Members
    • Member ID: 122,270
  • 392 posts
  • Gender:Male
  • Location:California

Posted 06 September 2013 - 02:36 PM

I had spikes of energy for iron, manganese, copper, cromium, lead, and uranium in both helmets.  The way the machine works is to shoot gamma rays to disturb the electrons on the element.  The energy that is put off when it goes back to it's electron shell is read by the machine as an x-ray value. 

 

Since I didn't pay for the machine I have tested pottery and toys for people for free, but had I forked out $28,000 for it I would probably charge what ever I could get away with.  I bet owners would still take $5 to $10 to provide that service.

 

Still everyone here is correct that there is variability in paint composition.  It could be from just not mixing the paint that day.  I will have to conduct a few tests on the fake helmets I have to see the differences.  I will probably not find lead on the fakes.

 

Brad



#30 Bill47

Bill47

    FORUM SUBJECT ADVISOR

  • FORUM SUBJECT ADVISOR
    • Member ID: 5,695
  • 555 posts
  • Gender:Male
  • Location:Massachusetts

Posted 06 September 2013 - 04:57 PM

So, can we put it more concisely by saying that while such testing may prove that a helmet is a fake, it can't prove that a helmet is authentic?  Or is that an oversimplification?



#31 Bugme

Bugme

    ADMINISTRATOR

  • Administrators
    • Member ID: 1,726
  • 15,260 posts
  • Gender:Male
  • Location:Marsaxlokk, Malta & Wisconsin Shoreline

Posted 06 September 2013 - 06:06 PM

So, can we put it more concisely by saying that while such testing may prove that a helmet is a fake, it can't prove that a helmet is authentic?  Or is that an oversimplification?


___________________________________________________
 
Yes for the first question and Yes for the second(for those who like complicated)



#32 jgawne

jgawne
  • Inactive
    • Member ID: 1,657
  • 3,048 posts
  • Gender:Male
  • Location:Outside Boston USA

Posted 07 September 2013 - 09:09 AM

I had a very long talk with one of the fellows who was involved with the German helmet testing service. What was the most interesting thing to me was the people against them were not as concerned about the potential financial losses they would suffer, but that their reputation for being a "helmet guru" -who could hold something and pronounce it good or bad - would suffer.

What the test does is tell you if elements that should not be in period paint are there (thus confirming a fake) or showing that the elements (especialy heavy metals) ARE there, thus ruling out a recent cheap reproduction. Hokum, it is not.

OF course it is possible for someone with the right knowledge and contacts to fake the paint, but this is why it is only ONE tool in the toolbox. And from everyone I have talked to it is a VERY good tool, but scared too many people, and was misunderstood by many more.

It essentially scans down to the bare metal, and you really have to have a database of different helmets to be able to figure out the correct percentages of paint ingredients. Painting over original paint with modern paint will just adjust the mix of elements, so this is where a good database to use as a baseline is important. Of course it cost a ton of money for them to fly around checking helmets of top collectors, and if they were to give up the database it would allow others to replicate their tests, AND provide info on how to better fool it. So the database was a proprietary thing. That they would not just give it away was the reason many claimed there was "no transparency" in the testing.

Of interest is that even with all the problems the Germans military had in production, the guys did find a very good baseline of what the readings should be. Just like the US the Germans also had specs to follow, and there was little variance.

But much of the tough issue is looking at over paintings and multiple layers. This, I think, was where the tester really had to know what they were doing to interpret the results.

The entire thing was financed by a guy with deep pockets, who unexpectedly died. He owns the database so it is tied up in his estate (a shame). As there was not really money to be made in it (consider equipment cost, the massive startup cost of the database, ongoing calibration and maintenance, etc.) this was NOT a cash cow. They started with a cost of $250 per helmet, but as they got better they dropped the price to $150.

They had just started to look at US helmets, and had gone through Mark Bando's collection of helmets- which all come directly from the vet, from before anyone cared about such things, and they felt there were getting good results. So from the people that had the most experience in the world doing this, they felt they could use the same technology on US pots- it's just few WW2 helmets were worth multiple thousands.

So, if anyone has a rich friend and wants to finance a project it appears that, yes, with enough samples a baseline could be created that would tell to a pretty good % if something used modern paint (which is probably 95% of the fakes) or if the paint in consistent with known period examples of paint.

I am now looking into how they use ultra high magnification to look at paint cracks, chips, and such.

#33 CherryPoint

CherryPoint
  • New Members
    • Member ID: 131,525
  • 5 posts

Posted 07 September 2013 - 11:52 PM

Gentlemen,

 

I started analyzing German helmets starting in 2009 with XRF and have scanned hundreds of helmets from some of the largest collections in the world resulting into a large database . XRF is a very powerful tool in providing the elemental composition of the object without damaging it. This includes the steel, paint, and decals. It is possible to match 5 layers  ( steel shell, 2 layers of paint, 2 set of decals ) on 2 different examples of German helmets with no problem. The elemental composition of all the different German decals from WW2 are also now known. 

 

In 2010, XRF analysis was started on some high end US helmets ( eg. WW2 Airborne helmets with provenance) and also provided some great data.  XRF will not tell you about the other components of paint that includes fillers and binders,etc. Other technology is necessary for this analysis

 

All the XRF analysis conducted was simply about understanding the elemental composition of  how these valuable artifacts were made and to address the fakers taking advantage of people.

 

Thanks

 

David



#34 jgawne

jgawne
  • Inactive
    • Member ID: 1,657
  • 3,048 posts
  • Gender:Male
  • Location:Outside Boston USA

Posted 12 September 2013 - 06:17 AM

I had a long talk with someone who felt XRF was pretty useless. However he was really unable to make a decent scientific claim to back it up. Again, from everything I have seen, pro and con, I still feel it has the potential for being a very good tool to help weed out 90% of fake helmets. Unfortunately, it is not as easy as pulling a trigger and getting a green or red light. It will take some work to build up a database, as well as learn how to properly read the results.

Possibly this will get cheaper and easier in the future, and then we shall see a lot of screaming by collectors who find some of the things they have are not what they seem. ("but the vet said he was wearing this when he shot Hitler!")

#35 Bugme

Bugme

    ADMINISTRATOR

  • Administrators
    • Member ID: 1,726
  • 15,260 posts
  • Gender:Male
  • Location:Marsaxlokk, Malta & Wisconsin Shoreline

Posted 12 September 2013 - 07:27 AM

Possibly this will get cheaper and easier in the future, and then we shall see a lot of screaming by collectors who find some of the things they have are not what they seem. ("but the vet said he was wearing this when he shot Hitler!")

The day will come where XRF or a viable alternative will come into use and as fast as technology improves, I'm sure that it won't be many years off. We're just not there yet. 

And you're right, there will be screaming. It's funny, when I started my helmet inspection service a few years back, using the best technology available at this time, I decided that I should be fair and test all my own helmets, which I was sure were all 100% legit. One of them turned out to be a fake, I tested it three times! How could a helmet, which my good eye could tell was legit... possibly be a fake!?!?! Well, I couldn't deny the results, my eye had been fooled. I now keep that helmet as a constant reminder that even the most experience and the best eye, can and will be fooled.



#36 Hürtgenwald

Hürtgenwald
  • Members
    • Member ID: 122,270
  • 392 posts
  • Gender:Male
  • Location:California

Posted 12 September 2013 - 10:31 AM

There are new XRF models that use a "bulb" instead of a radioactive source.  They are now more affordable and you don't have to be licensed with the Nuclear Radiation Commission to have the bulb type.  Used ones can be rented too.



#37 jgawne

jgawne
  • Inactive
    • Member ID: 1,657
  • 3,048 posts
  • Gender:Male
  • Location:Outside Boston USA

Posted 12 September 2013 - 10:41 AM

Actually, one individual who shall remain anonymous, told me the real problem with the XRF naysayers was not that they could be liable for thousands of dollars in "guaranteed" sales, but that they just did not want their 'reps' as masters of the helmets hurt. They were afraid people would no longer bring helmets to them for a blessing at shows, and thus their importance would collapse.

I find that funny as I would think the money aspect would be much scarier, but I guess not.

#38 CherryPoint

CherryPoint
  • New Members
    • Member ID: 131,525
  • 5 posts

Posted 15 September 2013 - 04:40 AM

IMHO, regarding US helmets , specifically the highly faked Airborne helmets. XRF is a good reference point but more data is necessary, esp with white paint  . The authentic  Beauchamp's http://www.ss-steel-...eel_feature.htm

 

helmet revealed some interesting XRF numbers. 


Edited by CherryPoint, 15 September 2013 - 04:44 AM.


#39 just-a-good-ole-boy

just-a-good-ole-boy
  • Members
    • Member ID: 2,533
  • 1,715 posts
  • Gender:Male
  • Location:The Deep South

Posted 15 September 2013 - 06:08 AM

How do we know that the Beauchamp helmet is in fact an original specimen. The paint on the spades and the helmet look somewhat odd to me, but I could just be a skeptic. I believe this one came from Ben Overhand if my memory serves me correctly. I have my own opinions about some helmets that have come from him.

 

Steve



#40 CherryPoint

CherryPoint
  • New Members
    • Member ID: 131,525
  • 5 posts

Posted 15 September 2013 - 10:30 AM

 Thanks to Steve ... Interesting...indeed.



#41 wildcat123

wildcat123
  • Members
    • Member ID: 1,095
  • 3,030 posts
  • Gender:Male
  • Location:Kansas

Posted 15 September 2013 - 10:55 AM

How do we know that the Beauchamp helmet is in fact an original specimen. The paint on the spades and the helmet look somewhat odd to me, but I could just be a skeptic. I believe this one came from Ben Overhand if my memory serves me correctly. I have my own opinions about some helmets that have come from him.

 

Steve

 

 

This is precisely the problem I tried to articulate in a previous post.  In spectroscopy, you have to have a standard to compare your specimen to.  You can build a library of different helmets from different periods of time so you can see the variability (most of us can determine what original McCord paint should look like, so not a big problem). Hurtgenwald show us the variability of just one element (lead) in one of his earlier posts while scanning different McCord helmet shells.  Decals should work fairly well as well, you have a set number of manufacturers that should end up looking similar as spectra.  On the high end helmets, it is not so cut and dry.   A person is going to have to say "Helmet A is good, and helmet B is bad" in order to build the needed library.  A good helmet should have X amount of lead in the paint, X amount of Element B, Y amount of element C, etc etc etc.  You would have to do this for both the McCord applied paint and the paint used to add any additional markings...

 

It is a simply a matter of arithmetic.  You cannot solve for 50 variables (variability within paint manufacturer, variability between paint manufacturers, variability in what paints were used, variability in how the paint was applied, how clean the brush/spray equipment was, etc, etc, etc) when you have 1 equation (the XRF scan).  Can it be used to weed some of the "crap" out of the population, sure.  Can it ever be 99% accurate, I do not believe that it can.

 

But you don't have to take my word for it...  but I'm not a 17 year old kid and my education did involve some Chemistry, haha.



#42 Bugme

Bugme

    ADMINISTRATOR

  • Administrators
    • Member ID: 1,726
  • 15,260 posts
  • Gender:Male
  • Location:Marsaxlokk, Malta & Wisconsin Shoreline

Posted 15 September 2013 - 01:03 PM

I think it's become pretty much established that there are way too many variables, at this time, to put our wholesale trust into a machine like XRF. Cherrypoint has not really established anything here other than to state that he has used it and found issues with US helmets. Sadly, several uS helmet were destroyed by Robbie Wilson after XRF determined they were fake. Yet, by Cherrypoints own admission the database wasn't very deep at all. While XRF has potential, it also has extreme limitations. I think that the last several posts address the elephant in the room, which is, who determines which data base entries are the "correct" entries? To that, there is no answer.



#43 anton67

anton67
  • Members
    • Member ID: 63,438
  • 5,723 posts
  • Gender:Male
  • Location:PA

Posted 15 September 2013 - 06:41 PM

Amen Bugme

#44 CherryPoint

CherryPoint
  • New Members
    • Member ID: 131,525
  • 5 posts

Posted 19 September 2013 - 03:59 PM

IMHO. I like to see Lead and some other elements in WW2 helmets...German and US...   so many fakes  

 

XRF is just  one tool .....interesting to see when one of the most valuable helmets in the world was discovered

 

http://www.ucl.ac.uk...icles/20100920b

 

The first thing they did was shoot it with XRF

 

 

S/F



#45 Bugme

Bugme

    ADMINISTRATOR

  • Administrators
    • Member ID: 1,726
  • 15,260 posts
  • Gender:Male
  • Location:Marsaxlokk, Malta & Wisconsin Shoreline

Posted 19 September 2013 - 04:06 PM

IMHO. I like to see Lead and some other elements in WW2 helmets...German and US...   so many fakes  

 

XRF is just  one tool .....interesting to see when one of the most valuable helmets in the world was discovered

 

http://www.ucl.ac.uk...icles/20100920b

 

The first thing they did was shoot it with XRF

 

 

S/F

However, they aren't dealing with multiple elements in paint, only the material from which this Roman helmet was made. I'm sure we could get consistent Hadfield Manganese marks with XRF on a line of unpainted M-1's.



#46 CherryPoint

CherryPoint
  • New Members
    • Member ID: 131,525
  • 5 posts

Posted 19 September 2013 - 04:25 PM

However, they aren't dealing with multiple elements in paint, only the material from which this Roman helmet was made. I'm sure we could get consistent Hadfield Manganese marks with XRF on a line of unpainted M-1's.

Paint signatures with elements are easily matched if analyzed from a manufacturer. This has been completed with several complex German examples..and extremely accurate  

 

I agree wide variables with paint such as white paint with unknown origins on US helmets are very difficult to match because we do not know  where the paint came from. But the XRF data is still useful for the database. 

 

However,  the fakers must understand the basic elements that are present in period WW2 paint.

 

Now... apply technology to understand the non elemental  data.. Without taking a sample.



#47 Bugme

Bugme

    ADMINISTRATOR

  • Administrators
    • Member ID: 1,726
  • 15,260 posts
  • Gender:Male
  • Location:Marsaxlokk, Malta & Wisconsin Shoreline

Posted 31 December 2013 - 05:00 PM

There is an incredibly lengthy discussion on this topic over on the K98k forum. While it is from the German helmet perspective, the information is priceless. Warning, it is a 67 page discussion: http://www.k98kforum...-and-Censorship



#48 Burning Hazard

Burning Hazard
  • Members
    • Member ID: 8,715
  • 2,272 posts
  • Gender:Male
  • Location:Canada

Posted 31 December 2013 - 05:43 PM

Interesting thread.

 

I have a XRF Spectroscopy machine at work (I'm an Oil Chemist by trade). Maybe I can ask my boss for permission to run ground up M1 helmet samples through it and see what pops up  :lol:



#49 Bugme

Bugme

    ADMINISTRATOR

  • Administrators
    • Member ID: 1,726
  • 15,260 posts
  • Gender:Male
  • Location:Marsaxlokk, Malta & Wisconsin Shoreline

Posted 06 October 2015 - 02:28 PM

Update: After speaking with a moderator(Hambone) on the K98 Forum where the discussion of XRF and Third Reich helmets has been ongoing for years, a new development has arisen which is really the death nail for XRF technology being used to authenticate helmets. While I won't get into a lengthy discussion about Third Reich helmets, some things need to be said relating to these helmets since it does apply also to those of us who collect U.S. helmets

TR helmets also have lot numbers which can be used to date the production and recent research has found that the "Champagne SS Runes" on a late war produced helmet are ALL fakes! (Every last $6000 to $24,000 one of them). These helmets never made it out of the factory much less into the field yet, the Champagne SS Rune marked helmets have long be touted as wartime used. XRF had found them to be legit and COA's were done to legitimize them. Now it comes out that everyone of them were done as postwar fakes that were done in the early 1970's. Several well known authorities among TR helmet collectors are now conceding this fact. As a result, XRF is being shunned like a bad rash by a great many on the TR collecting side of this.

Below is a response from one of the founders of XRFacts, (Maui on K98 Forum and Cherrypoint here on the USMF) in response to my question about U.S. helmets not being able to be scoped properly, lacking a database and too many large paint variations to be done properly. I posted his entire response to me here so you can see what kind of person we are dealing with. Anything written in Blue are notes added by me:
________________________________

[From Maui] Hello, I will address you once and once only ...and try to clear XRF up for you and BTW I am also a USMC veteran..Semper Fi and have some early USMC WW1 helmets . but primarily a German WW2 Helmet collector and I have quite a few ..... and my focus was not on US militaria per se .... and I don't care what you think or believe (He should care since his business will be based on people caring)

The vast majority of XRF testing was completed on the absolute prime, rarest and extensive WW2 German helmets collections around the world . Period ..the largest Database of German Helmets in the world ! Absolute consistency with all the manufacturers ..Steel shell. Paint..and decal ...of course this makes sense. And also some great examples of German Camo paint were tested and the understanding of certain key elements.

No one can dispute the findings in this arena..if you want to take the time like I did and rent a gun and get some german helmets to test ...do it and show the world !!! (We can now dispute the findings in this arena, the Champagne SS Runes which he declared 100% legit have now all been recognized as post war fakes)

You won't see anybody post any data refuting XRF data on German helmets period....... the energy spectrum is consistent with the German WW2 manufacturers.

Also XRF is only one tool ..remember...

(Then he makes this Admission about U.S. helmets:) Regarding US helmets ...basically uncharted from my perspective ..I have some data ...we shot several ...including airborne ...but white paint insignia is a dangerous minefield...XRF is not enough ..other technology is needed and I will leave it at that ...for example white paint zinc. titanium is available then and now..lead is another story and I don't care enough about all the different manufacturers..I have the data on some on the profiles of what are considered textbook airborne US helmets including ones that sold for very large amounts of money ...but not my interest...however I can say...a lot of fakes for sure ....

______________________________________

So, there you have it. These guys walked around the Show of Shows a few years ago handing out cards telling everyone that they were here to save the hobby. Sadly, all they accomplished was to create more problems for the hobby. We ended up having several rare and valuable U.S. helmets declared to be fake by XRFacts. These same helmets were considered by many seasoned collectors(including myself) to be the real deal. As a result, these helmets were marked and the owners were told they had bad goods because the gods of XRFacts had said so. Now with the most recent revelation concerning the Champagne SS Rune helmets coupled with the owners snarky admission to me that it doesn't work on U.S. painted helmets, I think we can surmise that the day of this flawed technology has now officially passed. I feel for the collectors who trusted XRFacts and destroyed or tossed perfectly legit helmets based on this.

I was originally interested in this "science" and was willing to give it an opportunity to be proven. Well, it has been proven to be fallible and will not receive endorsement by any of us here.



#50 wildcat123

wildcat123
  • Members
    • Member ID: 1,095
  • 3,030 posts
  • Gender:Male
  • Location:Kansas

Posted 06 October 2015 - 02:53 PM

Not flawed technology per se, it is the flawed application of the technology.  I won't beat a dead horse, I've already posted the problem with this technique when you don't have a baseline to compare to (basically finger printing).  Very interesting subject, nonetheless.




0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users