Coin Community Family of Web Sites
Coin Community Forum
Username:
Password:
Save Password
Forgot your Password?

Welcome Guest! Need help? Got a question? Inherit some coins?
Our coin forum is completely free! Register Now!

Testing US coins for silver content, part 2

Next Page    
To participate in the forum you must log in or register.
Author Previous Topic Topic Next Topic
Page: of 2
Pillar of the Community
Learn More...
5318 Posts
 Posted 01/14/2009  12:01 am Show Profile Bookmark this topic Add KurtS to your friends list Get a Link to this Message

Here's a new update in my series on testing coins for silver content now with an interesting twist. Ever since I noticed discrepancies in the density of silver coinage, suggesting lower silver content than mint specifications, I've wanted to know the accuracy of my testing methods. To that end, I initially tested a "control group" of bronze coins, supposing there was less economic motive to degrade a copper-based coin. Based on those tests, I attributed a margin of error at .8-1.5%

Of course, there is really no telling if these bronze coins were exactly to mint specs, except by additional testing. So what better way to determine accuracy than by testing something with a verified silver content, ie bullion?

Here is my new test subject, which should be very accurate for silver content, as well as a comparable weight to other test coins.
Any guesses at my test result? (.999 silver has a SG of 10.49)



My result (an average of 3 tests) is...10.49! (the variance between tests was 0.5-0.8%)
Now I think that puts my other test results in a very interesting light, wouldn't you say?

The chart below depicts the measured SG against calculations based on mint specifications, including a new test on a US coin.
The result for each sample below is the average of 3 separate tests. I have also noted baselines for .925, .800, and .500 silver content.



Here are the coins corresponding to the above chart: (discrepancy of SG vs mint spec)

1. 1853 25c = 10.11, SG test approx. .775 silver / .225 copper (-2.2%)
2. 1932 25c = 10.0, SG test approx. .750 silver / .250 copper (-3.3%)
3. 1829 50c = 10.18, SG test .800 silver / .200 copper (-1.7%)
4. 1954 Franklin 50c = 10.17, SG test .800 silver / .200 copper (-1.7%)
5. 1964 50c = 10.24, SG test approx. .850 silver / .165 copper (-1%, but getting close! )
6. 1964 50c (second test coin) = 10.22, SG test approx. .835 silver / .165 copper (-1.1%)
7. 1966 Mexico Peso = 9.06, SG test approx. .100 silver / .900 copper --this is as expected
8. India 1913 1 Rupee, SG tested close to .835 silver / .165 copper (-1.8%)
9. Australia 1954 Sixpence, SG test approx. .350 silver / .650 copper (-2.6%)
10. Switzerland - Frank. Two different coins, exact same SG results as .500 silver / .500 copper (it should be .835) (-5.7%)
11. Canada 1947 25c, SG tests closest to .750 silver / .250 copper (-1.1%)
12. Canada 1913 10c, SG test approx. .700 silver / .300 copper (-2.9%)
13. Newfoundland 1943 10c, Spec is .925, coin SG is closest to .500 silver / .500 copper (-6.8% )

SG for various silver alloys (where remainder is copper):

.999 = 10.49 (ASEs, silver rounds and bullion should test to this)
.935 = 10.41 (Some Spanish colonials)
.925 = 10.39 (Sterling silver, ie Canada to 1919, Australia to 1945)
.917 = 10.38 (Some Spanish colonials)
.900 = 10.31 (US Coin silver)
.850 = 10.23 (Swiss silver coinage)
.800 = 10.17 (Canada to 1967)
.750 = 10.08
.600 = 9.84
.500 = 9.68 (most Australian pre-decimal silver after 1945)
.400 = 9.53
.300 = 9.38
.200 = 9.23
.100 = 9.09 (Mexico billon Pesos 1957-67)

Pillar of the Community
United States
4000 Posts
 Posted 01/14/2009  01:44 am  Show Profile Check Jaobler's eBay Listings Bookmark this reply Add Jaobler to your friends list Get a Link to this Reply
Hi Kurt,
Your results are very interesting, but I'm having a hard time believing that US silver coins with a nominal 90% silver content are not very close to that value. Weren't coins tested every year by the Assay Commission to verify the metal content? Consider the scandals that ruined some of the companies making territorial gold coins during the Gold Rush; they were hounded out of business at the mere suggestion that their coins did not have full gold content.

If any of the US coins you tested were actually as low as 0.750 to 0.850 silver (75% to 85%) rather than the specified 90%, I think that fact would have come to light early on. This would have been an especially juicy scandal!

When trying my own specific gravity measurements I learned it is easy to get a low value. One major source of experimental error was that air bubbles were easily trapped beneath the coin or within the loops of the wire basket I used to suspend the coin. My own attempts did not convince me that you can rely on specific gravity to accurately determine silver content.

Pillar of the Community
Learn More...
5318 Posts
 Posted 01/14/2009  02:18 am  Show Profile Bookmark this reply Add KurtS to your friends list Get a Link to this Reply
Yep, I've been very skeptical too, but after hours of testing I'm more convinced by my results than official mint specs. Apparently a few are fairly close to spec in any case!
Now, if I had seen a consistently low reading across the board for coins of many compositions, that would have raised a flag to me.
As it turns out, my bronze and CuNi "control group" coins have tested very close to official specs, as well as several older coins of "pure copper" composition.
I know the effect air bubbles would have on density, and I have devised methods to reduce that error as much as I can.

Even if that were the case, it does not properly explain the discrepancy I see, particularly with the bullion bar. Let's just say the results are interesting--not conclusive, and I'll continue to test more coins. While my own tests have little weight, it might be an easy matter to send a few coins to a real lab for analysis.

Let's say for argument's sake my results have a semblance of accuracy, would that amount to a "juicy scandal"? I'm not convinced it would be considered a scandal (back then), particularly if those "holding the bag" didn't want that information public. How would people hear about it otherwise? Would they believe a "science geek" like me? I doubt it!
Pillar of the Community
United States
581 Posts
 Posted 01/14/2009  03:23 am  Show Profile Bookmark this reply Add Kabiye_Lady to your friends list Get a Link to this Reply
Hmmm. I played around with the math and it looks ok. About the only thing I can think of is wear. This comment is out of sheer ignorance, but as a coin wears, if you lose one gram of "coin", do you lose 0.9 grams of silver and 0.1 gram of copper? The other thing that points to wear is that your table shows as the coins get "newer" (and perhaps less worn?), the SG goes up slightly.

From a more practical standpoint. Do you really think the smelters and melting places haven't checked this out? I would think they can actually melt the darned things down and find out for sure. I don't think they would be paying for 90% silver and only getting 80%. That's probably the strongest argument.....

Another thought. A couple of years ago, I did a drive-buy purchase of a 1967 Canadian Half dollar. When I got home, I discovered that some were minted at 50% silver and others at 80% silver. As I researched this, everyone was saying there wasn't a non-destructive way to determine which coin (50% vs. 80%) you had. Why didn't anyone think of Specific Gravity?
Edited by Kabiye_Lady
01/14/2009 03:27 am
Pillar of the Community
Learn More...
5318 Posts
 Posted 01/14/2009  03:59 am  Show Profile Bookmark this reply Add KurtS to your friends list Get a Link to this Reply

Quote:
Do you really think the smelters and melting places haven't checked this out? I don't think they would be paying for 90% silver and only getting 80%. That's probably the strongest argument.....

To be honest, I don't know what to think--other than the results I've provided--interpret as you like. It's easier to test actual coins, than motives of individuals/companies. Of course, everyone out there is welcome to test a piece of silver bullion, pure nickel, or pure copper and compare the results to various coins. Regarding wear affecting density, my guess is the alloy doesn't differ much through the thickness of the coin, so Cu and Ag is worn off at an even rate. An exception of course would be heavily worn clad coins, which I haven't tested.

Once set up, it's not hard to perform these tests to compare to coins of known density. For instance, I just tested a 1978 Canadian 50c, which are 100% nickel and should have a consistent density of 8.9. Here, the average of 3 tests yields 8.84--which is pretty close to spec (-0.7%).
Pillar of the Community
United States
2535 Posts
 Posted 01/14/2009  09:25 am  Show Profile Bookmark this reply Add morgans dad to your friends list Get a Link to this Reply
KurtS, keep up the process of thought provoking exercises, great thread!!
Valued Member
United States
365 Posts
 Posted 01/14/2009  10:04 am  Show Profile Bookmark this reply Add SeriousCERES to your friends list Get a Link to this Reply
Hi all:
I've been following this thread with interest and I really appreciate the debate going on here.
I was intrigued by Jaobler's question about the potential for scandal: I have an easier time understanding why there was scandal and backlash around the circulation of poor-quality gold coins, gold, after all, having long been the backbone of the currency system. I have the impression that throughout the 19th century fairly low-grade silver circulated all over the US, before the US mint started making mass quantities of silver dollars-- absent, btw, from your tests so far, KurtS!--and so maybe some folks didn't look at silver with the same expectations as gold.
Of course, some perfectly metallurgical explanation may yet be forthcoming, so I'll hush and keep reading!
Pillar of the Community
Learn More...
5318 Posts
 Posted 01/14/2009  11:54 am  Show Profile Bookmark this reply Add KurtS to your friends list Get a Link to this Reply
I wish I had some Morgan dollars to test, as well as gold coins.
I may borrow some from a friend, just for curiosity's sake.
Valued Member
United States
178 Posts
 Posted 01/14/2009  12:06 pm  Show Profile Bookmark this reply Add Bronxman95 to your friends list Get a Link to this Reply
How bout you see if the US govt is being honest and test a .999 Silver Eagle?
Pillar of the Community
United States
1815 Posts
 Posted 01/14/2009  1:21 pm  Show Profile Bookmark this reply Add steve199 to your friends list Get a Link to this Reply
Kurt,

That is some interesting stuff. I have thought trying some home SG testing to check out metal content...knowing that I would be testing myself more than the coins. :)



Quote:

Another thought. A couple of years ago, I did a drive-buy purchase of a 1967 Canadian Half dollar. When I got home, I discovered that some were minted at 50% silver and others at 80% silver. As I researched this, everyone was saying there wasn't a non-destructive way to determine which coin (50% vs. 80%) you had. Why didn't anyone think of Specific Gravity?


Kabiye, if it is a half-dollar, it should be 80% silver. The quarters and dimes switched to 50% silver during mid-1967. Then to keep everyone on their toes, switched mid-stream to no silver in those two coins in 1968. Someone with the ability to make very accurate measurements should be able to non-destructively tell the difference...but the smaller the coin, the more accurate the measurements have to be. A dime doesn't displace very much water.

Pillar of the Community
Learn More...
5318 Posts
 Posted 01/14/2009  1:40 pm  Show Profile Bookmark this reply Add KurtS to your friends list Get a Link to this Reply

Quote:
...knowing that I would be testing myself more than the coins.

I viewed this exercise as a challenge to myself--and sought to quantify how much error my methods introduced, and whether anything could be interpreted from my results.
Maybe it suggests something, but I'll keep on testing when I get new coins.

I should mention that I use a fairly accurate jeweler's scale (+/- 0.01 gram) which has tracked well to actual lab results on gemstones I own.
My method for testing SG follows Swamperbob's post here. What's nice about this method is it doesn't attempt to measure water, only the coin.
Pillar of the Community
United States
1815 Posts
 Posted 01/14/2009  2:01 pm  Show Profile Bookmark this reply Add steve199 to your friends list Get a Link to this Reply
Cool...thanks for the link. I had missed your post from Christmas day where you started this whole process.


Pillar of the Community
United States
1815 Posts
 Posted 01/14/2009  4:39 pm  Show Profile Bookmark this reply Add steve199 to your friends list Get a Link to this Reply
Hello again, Kurt,

It looks like the scale is hanging off the edge of the desk. I'm sure you checked this...the scale actually gives an accurate measurement like that?

Edited by steve199
01/14/2009 4:40 pm
Pillar of the Community
Learn More...
5318 Posts
 Posted 01/14/2009  5:07 pm  Show Profile Bookmark this reply Add KurtS to your friends list Get a Link to this Reply
Steve,
Yes, the process by which I weigh a coin suspended varies little from weighing the coin normally on top of the platform. Naturally, every scale may differ in this respect, and here I am particularly careful to ensure reasonable accuracy. Obviously, nobody else can prove/disprove the accuracy of my own methods, so I suggest they work from Swamperbob's explanation using a good scale of their own, starting with a control sample of known compositions such as pure copper, nickel, and silver. Then, if after repeated tests an accurate methodology is established, continue on to coins of uncertain composition. And keep testing control samples at periodic intervals to ensure your methods haven't slipped (or the scale hasn't decalibrated). That's simply the approach I've taken to get my results.

Let me stress here that I'm hardly an expert doing this--I'm just following the method linked above with my own equipment as accurately as I can. If you want to discuss methods further, feel free to post to my demo thread where this is discussed in more detail. Thanks!
Pillar of the Community
United States
1815 Posts
 Posted 01/14/2009  5:25 pm  Show Profile Bookmark this reply Add steve199 to your friends list Get a Link to this Reply
Kurt, I was just asking if the scale is sitting flat on the desk...or if it has two feet hanging off. If the feet are hanging off, it somehow is not affecting the accuracy.

I need to avoid get sucked into this, or I'll be at the house with a mad scientist hat on my head.
Pillar of the Community
Learn More...
5318 Posts
 Posted 01/14/2009  5:57 pm  Show Profile Bookmark this reply Add KurtS to your friends list Get a Link to this Reply
Well, the first thread shows the scale hangs over the edge, and I just followed up to comment here on accuracy.
As shown in that photo, the harness focuses the weight close enough to the sensor under the stage; the feet aren't a critical part of measurement, as indicated by tests.
Mad scientist or not--I found these tests very interesting.
Page: of 2 Previous Topic Topic Next Topic  
Next Page    
To participate in the forum you must log in or register.
Coins Commemorating the 2014 Sochi Winter Olympic GamesCoinSniper.com is the numismatic penny auction.SWCoins - We Love CoinsSELECTED Modern WORLD Coins with LOW MintageAmerican Silver Eagles only $2.49 over Spot!
Coin Community Member eBay Sales - Only CCF Members Included Here
Ending Soon   Newly Listed   Lowest Price   Highest Price  
Certified Coins   Certified VAMs   Certified Errors  
New Forum Topics Recently Active Forum Topics




Disclaimer: While a tremendous amount of effort goes into ensuring the accuracy of the information contained in this site, Coin Community assumes no liability for errors. Copyright 2005 - 2014 Coin Community Family- all rights reserved worldwide. Use of any images or content on this website without prior written permission of Coin Community or the original lender is strictly prohibited.
Contact Us  |  Advertise Here  |  Privacy Policy / Terms of Use
Coin Community Forum © 2005 - 2014 Coin Community Forums Go To Top Of Page
It took 3.47 seconds to rattle this change. Powered By: Snitz Forums 2000 Version 3.4.05