Keep in mind "LIBERTY" is an incused design feature of the nickel and based on my analysis of all four (Normal "B" business strike, Proof, Broken "B" business and Proof) the position of the date is static.
Yes. It takes several hubbings to create a working die, which is how we can get design hub doubling. An unfinished 1908 obverse die could only be made into a 1909 die by being hubbed with a 1909 working hub, as there's no date punch anymore. The effect on LIBERTY clearly shows multiple designs. This happened on 1878 Morgan dollars, too, as they changed the obverse design a couple times during the year.
@KBBPLL . i have started my collection. last night at Heritage I won a 1906-D barber dime in AU. I will have to get the coin in hand but i am hoping it is FS-10-1906-D -303 (10C-010.089) from the Cherrypickers guide.
Impossible to tell from that image, date position looks the same at least. Let us know when it arrives. Here's an MS65 with a great resolution image to compare. https://coins.ha.com/itm/barber-dim.../60349-50085.s?ic4=ListView-Thumbnail-071515#
I'm not sure if I get what you're saying. My understanding is that there was a design master hub, dateless, reduced from the galvano. This was then used to press a master die. That master die was then used to press one or more working hubs, which went on to be hubbed to however many working dies were needed. The master die could have been used across multiple years to create working hubs, as long as it remained in good condition, as there was nothing different from one year to the next. (I have no idea if they actually made a new master die each year prior to 1909 or not.) My understanding is also that they did not use the design master hub to directly create working dies. They could have, but if that became damaged they would have to reduce another one from the galvano. Prior to 1909 nickels, the date was then punched into each working die, as is evidenced by multiple date positions within each year. Starting in 1909, as pointed out by @justafarmer, the date position was static within each year, but moved from one year to the next. To me this implies that starting in 1909, a new master die was created each year, and the date was punched into that, instead of each working die. Each hub then had the date in the same position, and all the working dies were also the same. However, I think they were also creating a dated master hub from the dated master die, and all the working dies descended from that one hub. My theory on the 1909 broken beer belly B (in a nutshell hopefully) is that Liberty was damaged on the master die created for 1909. Keep in mind that Liberty would be a raised feature on this die. Maybe it was poorly hubbed to begin with, or the damage occurred on the dated master hub when pressing it. A working die (or dies) was then created for both circulation and proof coins, and a small number of coins were struck with the damaged B. This damage was noticed, and Barber simply re-engraved the master working hub with a cleaner Liberty, which now had the "regular B". The rest of the 1909 circulation and proof dies and coins descended from that. Keep in mind that the vast majority of 1909 nickels have the "regular B." The "regular B" disappeared in 1910 and all subsequent years because they simply made a new master die from the original master design hub, punched in the date, and went on about their business. Even though in my opinion the "regular B" is better aesthetically than the beer belly B, it was only re-engraved into that one 1909 dated master hub, and wasn't worth duplicating on the overall master hub. @justafarmer and I disagree on the circulation versus proof Broken B coins, because I feel that it's unlikely that a nearly identical broken B happened from two different things.
@KBBPLL . I agree about the pics. I just figure I only over paid by $30 dollars so I didn't have much to lose. Also, this is a close as i get gambling,. james
The thing that got me started with design hub doubling is actually the I in LIBERTY, specifically the slant of the bottom edge of the base. The proof coin on the bottom right shows a smaller, slenderer I with the base sloping down and to the right within a large I with a more level, but jagged base. This is the appearance I would expect if an undated, unfinished 1908 die were to be re-hubbed with the 1909 hub, which is the only way to get the date on that otherwise perfectly usable die. I guess if there were dated 1909 working hubs with both designs, it wouldn't have to be an unfinished 1908 die. The upper loops of the B wouldn't line up correctly, leaving the broken B appearance. The inside of the upper loop, being raised on the hub, would flatten part of the existing upper loop (raised on the die) of the beer belly B, depending on how many hubbings were in the die already from that hub and how many were given with the other hub. Do we know when the 1909 proof nickels were made? Given the relatively high mintage, I assume it was in at least two batches, one to go along with the proof sets being ordered at the beginning of the year with Indian cents, and again to fill orders after the Lincoln cent was struck.
Just to add a little more info to @KBBPLL post following is a picture of a 1883 No Cents nickel. Notice it also exhibits the Beer Belly "B".
Actually messy you pointed out the characteristics on the proof where our disagreement originates. @KBBPLL - I was avoiding a reply hoping you would cause I am out of BC Powders at the moment.
Kind of glad I started playing with this thread. I've just added the proof with what I'm calling design hub doubling to my want list. It'll make an interesting type coin.
I think I kind of get what you're saying now, but I don't think there was any separate design that could have been doubled. The "regular B" didn't occur in any other year. But as you say, the B might have been damaged by subsequent pressings of the usual beer belly B being slightly offset. I don't have a good explanation for what caused the slight "Longacre doubling" seen on some proof examples, but it appears to me that the "regular B" letters have filled in those areas. Here is an image of a 1908, 1909 UNC broken B, 1909 proof broken B, and 1909 proof regular B. I'm sorry I didn't get the last image quite the same size. If you zoom in, you can see that there's a "ghost" image on the proof broken B (3rd image) for example at the upper left corners of E and R. On the regular B (4th image), this has become a knob at those corners. You can see this also on the L, the strongly struck portion of which is the same as the 1908 L, and the "ghost" becomes the regular B version, where the lower right tip is more squared off instead of curved. So I don't know what caused this, but as I said above I think this dated master hub was just re-engraved to produce the regular B. I don't believe there was ever any overall master design with the regular B. I enjoy speculating about this stuff and getting other opinions. I don't know when the proof coins were struck but I think it was standard practice to do a run of circulation coins before the proofs.
And for fun, here are my collages. The 5 circulation broken B I was able to find. And the 7 broken B proofs I could find.
@KBBPLL . Yes, I used that photo but there is a large blow up pic on page 46 in the cherrypickers guide book. 4th edition, Volume II. I had to of course max the picture to 125% and then use a magnifying glass to identify dates protruding from the denticles. I was able to find only 4 though and the book mentions 9. James
Remember that the rules for appearance of doubled dies on incuse design elements is reversed from what it is for raised design elements. The "smashed" look that is normally called machine doubling is a sign of a true doubled die for incused elements, while notched incused elements are machine doubling. I'll check with John Dannreuther to see if he has that info. He recently released his book on proof nickel coinage. I just checked the book and he missed this anomaly for 1909. He only has one die pair identified, but that's probably due to the date position being fixed. He does say in the text that he didn't go looking for markers that would indicate multiple working dies. The photo he uses in the book is of a coin with a regular B. I sent a link to this thread to him, so we'll see if he has any comment.
For those who may be following the thread and are somewhat confused following is an image with arrows pointing to the characteristics of the Normal "B" displayed by the Broken "B" proof @messydesk is referring to.
@KBBPLL . Could you please clarify for me which of the two 1901-O types is the one to search for. The type 1 or the type 2? also, are there any population reports that are associated with these coin types. James