Purple Heart Perf Error

A local non-profit gives me the stamped corners off their incoming mail. Thus I am able to get a good number of current used stamps for my collection, but don't know the source.

I came across this odd stamp.



It is clearly larger than the normal variety, but has the odd perf skip in the middle. since the pers do not match up from one side to the other, it is not simply damaged perfs. It was cut this way, and this added section made the stamp larger. I am wondering how this happened.

*** Moved by Staff to a more appropriate forum. ***

That looks very unusual. I wished you could have scanned it - the photograph is distorting the measurements.
One thing is very odd. The flat part in the middle looks like it should have been at the corners, but the corners look normal. Also the die cut is peak/valley on one side of the flat part, and exactly opposite ( valley/peak ) on the other side.
Shifts like this occur on coil stamps fairly often but not quite this dramatic.

Peter

I seem to remember having read about this stamp somewhere. Perhaps in U.S. Stamp News (Modern Varieties column). Will see if I can locate the issue & refresh my memory.
Added: This stamp was discussed in the February 2016 issue of US Stamp News ... page 15. The stamp apparently appears in the leftmost column of some panes of Scott 4704b. The theoretical explanation is that the die-cutting mat was accidentally shifted to the right by the width of one stamp. Not exactly sure how this supposedly produces what you see ... will have to reread the article a few times & digest what it has to say. What I do understand though is that the separators that normally appear between adjacent stamps in a row somehow get moved to the center of the leftmost stamp in the row.
Peter: You are correct, I did mention this stamp in a previous thread. The Feb. article was apparently a follow-up to the October 2015 article (also in USSN) that first introduced the anomaly.

JLLebbert, "Are they rare PNC's" is the name of the thread in which you mentioned the above (freak?) (error?) (oddity?). In the thread you mention an article in U.S. Stamp News, in a column about modern varieties. I wish you could find that article!

Peter

It is actually half a stamp. Oops. Yes a whole stamp.

That is because the stamps were printed in larger panes with gutters between. The die cuts in the gutter have the same long space, but half the width of the stamp. See image below. If the cuts are shifted to the right, you get Alub's stamp.

Some might (or did) find the left most pane, which would be half imperf I presume.

Thank you for the thoughtful comments.

If I understand what you are saying, the horizontal and vertical die cuts are done separately. Otherwise, my stamp would have a straight edge and be half-sized.

But what does not make sense is how the corners came out perfectly.

Here are a couple of extra photos.







This stamp is a bout 0.5 mm wider than a normal copy.

Alub, that means the gutter is a bit wider than the stamps then. Horizontal and vertical cuts are done together in one operation.

I believe part of the gutter is now the left side of the stamp - am I correct?

Peter

BeeSee: Thanks for making the explanation clear enough for even me to understand. While you used 2012 Purple Hearts (4704) to make your case, the same would hold true for the 2014 ones (4704b). Since these were produced in 3x7 sheets, there would be 18 panes whose leftmost columns contained the weird die-cuts ... and 3 panes with partially imperf stamps in the left column. The article to which I referred posits that the panes with imperfs may have been caught & destroyed.

JLLebbert, I did not notice the date. I borrowed the image from ebay, which I will of course return later .

JLL, I am lost with the arithmetic. Can you explain?

Peter

Peter, I agree with you that the die-cuts that normally appear in the gutter now appear in the left most column of each pane. As for the numbers, I was simply pointing out that these were produced in 3x7 sheets (21 panes). That's 3 rows of 7 columns of panes. So each sheet would have 18 panes whose leftmost column should contain the strange die-cuts. The leftpost 3 panes would, as BeeSee suggests, contain partially imperf stamps in their leftmost columns. As no imperfs have been found ... and since partially imperf stamps would be much more easily recognized than the strange die-cuts in one column of a pane ... I agree with the author's idea that they were probably identified & discarded. Of course, the placement of the die-cutting mats may well have been realized & remedied. In fact, the discovery of a few imperf panes would likely lead to the discovery of the problem well before all the stamps had been printed. So there's no way of really knowing how many of these strange die-cut stamps exist. We know that there are 4 stamps in each "error" pane, 18 panes with the errors in each press sheet ... hence 72 "errors" per sheet. But there's no way to know the percentage of sheets with errors that made it through the production process.

OK, now I understand it much better. Thank you sir!

Peter

Ok, I get it now. That weird perf area is where the panes were designed to be cut apart. Mine must be a stamp on the edge of the pane, as the rest of the stamps would look fine. Of course, it is not possible to tell which direction the shift was, so we cannot tell which side of the pane my stamp came from. Thank you.

Any idea how many of these are out there?

thanks

Joe

alub: The USSN articles showed examples of the errors within blocks of 4. The stamps with the error were all from the leftmost column of a pane. Assuming that the die-cutting mat theory is likely correct, this would be the case for all such "errors". As I mentioned previously, the stamps were produced in large sheets of 21 panes. For those sheets in which the mat was mispositioned, 18 panes would contain the errors while 3 panes would contain partially imperf stamps. The latter were likely discovered & destroyed.

Folks, I don't understand the conclusion that there should have been some half imperf stamps. It depends on whether the stamps were aligned so that the direction of web travel was left to right or top to bottom relative to the designs and whether the direction of the die cut shift was parallel to the web travel direction or perpendicular. While you would get half imperfs if the shift were perpendicular to the web travel, you would not get half imperfs if the shift were parallel to the web travel.