"Longer Die Life"

[Robert]

Aug 25 Coin World has an article on the Mint's current experiments on using coatings to extend die life. They're toying with "Particle Vapor Deposition" and "Diamond-Like Carbide" coatings.

The article says that Lincoln Cent dies can last up to 1 million circulation strikes while nickels and dimes are more like half a million. State quarter reverses are about 250,000 and proof statehood quarters are about 1200 strikes.

I'll try to find more about PVD and DLC coatings.

[Robert]

As near as I can tell...




PVD: http://www.sulzermetco.com/eprise/Sulzermetco/Sites/Products/OtherProcesses/pvd.html

Sulzer-Metco is a well-known coatings company. It sounds a lot like "chemical vapor deposition" (CVD) where they put the part to be coated (base metal) in a vacuum chamber, pump out the air (oxygen) then heat a chunk of coating material until it "sputters" or splatters little bits of itself on the base metal, forming a coating. The coated part is then taken out of the chamber and heat-treated to make the coating stick really tightly.



Info on DLC:
It's basically a similar process except that it produces a low-friciton coating. The low temperature operation of this process also lends itself to coinage dies because there is low distortion due to heat.


http://www.diamonex.com/protectivecoatings.html

http://archive.metalformingmagazine.com/1998/06/DLC/dlc.htm

[Art]

I read the same article and a question on the low strike counts for state quarter proofs sprung to mind. They said about 1200 strikes from a die. They also stated that all of the proofs are struck twice. So does this net out to 600 quarters from each die before it is retired? Given the number of proofs being made, this would be a fantastic number of dies.

Does anyone know if this info and/or my inference are correct? Even if I'm incorrect, 1200 quarters to a die seems off by about 100x. I'd be far more prone to guess 100,000 + quarters (not strikes) to each die for the proofs.

[Bob P]

Art,
The numbers given for proof dies on the Statehood Quarters seems accurate. Since they are proof dies, and do strike the coin twice, then between 1200-2000 coins per die is about right. As soon as the proof dies begin to show wear to the point that the coins appearance suffers, they are removed from service and replaced. Proof coins are made mainly for collectors, so the mint tries to keep them as close to perfect as possible. You must also remember that the average number of proof coins minted for the first two years of the quarters program was 800,000, so not a lot of dies were required there.
As far as business strikes go, I have seen quarters that look like the dies were used for way too long. It is not impossible or uncommon for a die to strike 1 million coins, but I would venture to say that those coins at the end of the run look pretty bad. I hope this helps you out some.

Bob P

Bob P

[Robert]

The article does say an average of 1200 strikes per die, and that each coin is struck twice. This is an average of both copper-nickel and silver proof dies.

So it does appear that each proof die covers 600 coins.

So with a conservative estimate of 2 million clad proofs for the "regular" 1997 mintage, that corresponds to over 3000 proof dies. No wonder proof sets are so expensive!

And no wonder they're looking for better coatings for their dies!

[Bob P]

I am sorry to disagree with the "only 600 coins per proof die" theory. According to another Coin World article from 31 December 2002, and additional information submitted directly by the US Mint concerning die life, the numbers are much more. The minimum 'average' amount of strikes/hits from any proof die was 1,902 for a 1999 Delaware reverse. The most average strikes/hits was 5,431 for the Washington Obverse. At these rates, with two strikes per coin, then the average would be closer to the 1200-2000 I had mentioned earlier. Somewhere along the line, the CoinWord people got conflicting information. I would tend to agree more with the numbers sent directly from the US Mint. The whole article, including this exact topic can be found at this link:

http://www.coinworld.com/News/011303/News-1.asp

Please take a look at the last table in the article for proof coinage info.

Bob P

[Art]

Good information and an interesting article. Thank you.

I believe the numbers from the mint must reflect coins struck per die vs. the number of strikes. So to produce 2000 quarters the die would acutally hit the coin surface 4000 times.

So my next question then becomes, "Why are the number of coins (strikes) per proof die rising so drastically? Is there an improvement in design of the coins? Or are there just fewer top quality proofs?"

[Bob P]

I think some of your question Art were running through my head as well. The Mint chooses to use the word "Hits" when referring to the die striking the coin. Probably just a word thing, but interesting none-the-less.
As far as the increase in the amount of strikes each die makes, that can be explained by a couple of things. The design elements, if more intricate, tend to decrease die life because of the possibilities of cracking between close elements. The metal composition of the planchets are the same, so we can discount that as a reason. Getting the striking pressure perfect is probably the biggest factor, but then again, that is only a guess on my part.
As far as the proofs being of poorer quality, I don't believe so. The ones I have seen are very nice indeed. I would however, like to be a fly on the wall at the mint someday, and watch them go through an entire die life for a proof coin. I would be curious as to what point the strike is not of the quality that the mint wants to put out. Another question would be how often do they check the coins to see how they are coming out. The numbers they give are specific, not general in nature. Interesting all right.
I guess it's time for me to visit the mint and get a tour. Do you think they'll let me in there if I show them my library card?

Bob P

[Art]

So if the die produces say 1800-2100 coins and the presses turn out coins at the rate of 300 per minute (pretty slow), a new die must be inserted every 6 to 7 minutes.

I certainly would like to see this process in action.