Numbering machines: some thoughts

News: Welcome to the Canadian Paper Money Forum.

The process of printing the serial numbers on Canadian notes has always been separate from the printing of the rest of the note design.  That means different equipment is used for the printing of serial numbers.

Has anyone noticed that the style and size of numbers and letters on the multicolour notes (1969-1980s) is the same as on the Bird series notes (1986-2000s)?  The multicolour notes have red and blue serial numbers while the Bird series notes have black serial numbers, but the font and size of the digits is exactly the same.  The type and colour of ink used does not require a change of equipment between series.  So I have to believe the same numbering machines were used to print serial numbers on both series.

As well, the Journey series notes have the same style and size of numbers as the short-lived 1979 series $5 and $20 notes.  Did anyone else notice that?  The 1979 series was supposed to be a radical departure from conventional numbering practices, and it was.  The numbering machines used for the multicolour series had only 10 reels to accommodate a 7-digit serial number and either a 2-letter prefix (with or without an asterisk (*) in front) or a 3-letter prefix.  The numbering machine for the 1979 series had 11 reels to produce a sequence of numbers that was meant to be machine-readable, or so the theory goes.  For a number of years in the 1980s, the bank note printers evidently used two types of numbering machines for different denominations, and it's enigmatic that the Bank of Canada went back to the older technology for the Bird series, but evidently they did.  Perhaps they realized the old machines with whatever limitations existed still suited their numbering practices.

The old numbering practices dating back to the multicolour series continue to be used to this day (7-digit serial number, 3-letter prefix).  However, for the Journey series and the current Frontiers polymer series, it looks like the printers are using the numbering machine designed for the 1979 series.  There's nothing inherently wrong with using a machine capable of printing 11 characters to print only 10 characters.  It just means 1 reel never gets used.  And we know that there are no limitations on what-characters-can-be-printed-where because hundreds of millions of consecutively-numbered $20 notes of the 1979 series were made.  Maybe it was easy to adapt the machine from one format using only numbers to another using numbers and letters.  The extra reel is there but it never prints anything.  Maybe some day, we'll have notes with a 4-letter prefix or an 8-digit American-size serial number, who knows?  Apart from the similarities in size and style of characters between the 1979, Journey and Frontier series, there's another bit of evidence that just jumped into my mind to make me believe the printers are using the same numbering machine for all these series...

{http://wiki.cdnpapermoney.com/images/3/36/%2420_ARB_5339808_F.jpg}

A few collectors have found Journey series error notes with mismatched serial numbers where a black rectangle appears immediately to the right of one of the serial numbers.  If you ask me, that's evidence of the 11th reel, the one that is always supposed to be blank.  Well, in a machine malfunction, it can rotate too, though why a black rectangle appears, and not something else, is a mystery.  As far as I know, nobody has found an error note with an extra character like a number or letter, indicating that these characters exist on the 11th reel.  If such errors were ever made, they would be quite rare and would almost certainly have been noticed by the machine operators or quality control inspectors.  Notes with black rectangles are exceedingly rare.

As far as I know, this subject has never been discussed before.  Did someone beat me to the punch and publish similar hypotheses elsewhere?  And what do you think of the idea of old printing technology being used to make our modern plastic notes?

Here is an image of a 1979 series note for comparison purposes:

{http://wiki.cdnpapermoney.com/images/1/1b/BC-53aT_33000137347_R.jpg}

Great read and insight Thanks

It is possible that we will eventually go back to the 1979 serial number system... i had always wondered why they switched over to that system (11 digit numbers) for just those two denominations -the 1979 $5 and 20. Why did not continue this on the birds series? I understand that they were "experimental" but does anyone know why the Bank of Canada cancelled that serial number system?

It is possible that we will eventually go back to the 1979 serial number system... i had always wondered why they switched over to that system (11 digit numbers) for just those two denominations -the 1979 $5 and 20. Why did not continue this on the birds series?

Charlton (p. 281) says that "With the imminent exhaustion of the possible prefix letter combinations, the triple letter prefix was introduced in 1981." They needed to do something with the serial number, and with the Birds series likely in
early development stages by then, it was likely considered too expensive to do a short-lived 1979-style redesign with a new series just on the horizon. The triple-letter prefix wouldn't have cost much to implement compared with the a redesign, so that's my guess as to why they went that route.

By the time the Birds series came out I guess the triple letter prefix was meeting their needs so they stuck with it. I'm glad they did too. I never bought into the idea that serial number 56234444444 could be a solid radar.

The process of printing the serial numbers on Canadian notes has always been separate from the printing of the rest of the note design.  That means different equipment is used for the printing of serial numbers.

If you ask me, that's evidence of the 11th reel, the one that is always supposed to be blank.  Well, in a machine malfunction, it can rotate too, though why a black rectangle appears, and not something else, is a mystery. 

The 'Art and Design" of Canadian Bank Notes provides a photo of a numbering machine on page 9.

In reading about the production of banknotes it is clear that there are at least three different methods of printing used in the production of one banknote with each method separate from the  other. The methods  use different plates and possibly different machines: intaglio (for the raised portions), lithography (for all the background) and letterpress (for the numbers). Each step is separate and involves multiple plates. The book indicates that there are up to 8 plates used for the background printing alone (front and back).

Also looking at the machine one sees only 10 placeholders for prefixes and numbers. Although it would be easy to add in a spindle for the 11th placeholder I suppose.

Is there anything published about this 10 v/s 11 and continuation of use of same machines or is it purely a hypothesis at this time? The photo of the machine does not seem to support the hypothesis unless I am seeing it wrong.

I would hope that the BOC would be friendly enough to satisfy this aspect of curiosity given that it published such a detailed book on design in 2006.

Why did not continue this on the birds series? I understand that they were "experimental" but does anyone know why the Bank of Canada cancelled that serial number system?

If the last 7 digits are for serial numbers and the first 4 numbers or 3 alphabets are for denoting possible combinations of printer, denomination etc. does the simple combination of 10^4 v/s 26^4 imply that 3 alpha prefix give more possibilities than 4 numeric possibilities?

Of course one does need to account for the inability to use certain letters in prefixes and certain numbers in the 4 positions.

For about 15 years, the 1979 series was produced at the same time as the multicolour and (subsequent) birds series.  The printers could not have been using one numbering machine to handle all of these series at the same time.  The logistics of doing so are unthinkable.  The abandonment of the 1979 series probably had nothing to do with the birds series.  I would think that if the printers were using just one numbering machine, the 1979 series $20 note would not have been produced for 15 years.  They would likely have rolled out the Bird series $20 note much sooner.  But having the second numbering machine allowed the 1979 series to keep rolling, thus putting less pressure to get the Bird series released.

The Bank note printers did not print notes just for Canada.  They did work for all kinds of businesses and other countries, so having the extra equipment does not seem like a logistical fallacy.  Look at Canadian Tire money... that appears to be the same numbering machine producing serial numbers for CT money as for our circulating legal tender. And CT money has a 10-digit number instead of a three-letter prefix and seven-letter serial.  I would think that with a machine with a capacity for 11 reels, you can print with any number of active reels up to 11.  Maybe the actual machine used for the 1979 series is not the same one being used today, but perhaps the reels were pulled from it and put into a newer machine with only room for 10 reels.  I'm sure they swap out reels all the time as reels get worn down and/or broken.  If anyone has a better theory what the black rectangle is on a few of those mismatched number error notes, I would like to hear it.

As for the collectibility of the 1979 series, the argument of 7-digit radars not being true radars is pretty much on the same plane as saying that modern insert notes are not really replacement notes because they look like regular notes.

Well, I still think THIS number is pretty cool...

{http://img163.imageshack.us/img163/9207/56222222222s.jpg:http://img163.imageshack.us/img163/9207/56222222222s.th.jpg}

For about 15 years, the 1979 series was produced at the same time as the multicolour and (subsequent) birds series.  The printers could not have been using one numbering machine to handle all of these series at the same time.  The logistics of doing so are unthinkable. 

Perhaps there needs to be a distinction being made between "one numbering machine" and many instances of "one type of numbering machine".

If one looks at the picture given in the "art and design" one  can see how that machine can be readily fitted to print 11 digits instead of 10 without changing the whole process of printing or the machine. Presumably when different things are being printed they would be using a different instance of the same type of machine with the wheel set to print 11 positions rather than 10 or 9 for that matter.

It should be noted that the digits and characters are separate print heads and typically interchangeable on these machines so I do think that the notes having similar fonts has much to do about design decisions rather than printing issues.

Here is a photo of a conceptually similar machine:

http://www.gi-de.com/gd_media/media/press/bilddatenbank/GD_web-437_LIGHTBOX_LANDSCAPE.jpg

In such a machine the numbering assembly can be changed to hold whatever number of digits one needs to print.

So if there were runs of the Scenes of Canada original series (original multicolour) or the 1979 versions of the $20 and $5 of the same series or the Birds of Canada series the same basic type of machine could be used with simple modification to the printing head.

Based on their requirements the printing company could deploy various strategies of printing different types of notes including: scheduling, more pressess etc. This would be a business decision of the printing company and not of the BOC or the printing company clients.

As to how the number misprinting and the rectangle could have happened a possible scenario could be as follows:

Lets say a press was set up to print the 1979 version of $20 multicoloured. Lets say that the printing company decided to use that set up to do some trial runs for the $20 journey notes. Since in this case the 11th position would be "live" they decided to set it to a blank (similar to a space) rather than remove it completely. Erroneously they set it to a rectangle instead of a blank. They printed some notes.

In terms of how these errors got out:

Possible oversight.

Will this happen in any notes after the 1979 $20 and $5 were discontinued? Very highly unlikely because the printing company would have taken out the 11th position completely as it would not be needed.

My humble opinion :-)

Have you noticed that the font used in your passport (at least the ones before the artwork on the pages - pre 2012 maybe) is the same as the font used on Journey and frontier notes?

Have you noticed that the font used in your passport (at least the ones before the artwork on the pages - pre 2012 maybe) is the same as the font used on Journey and frontier notes?

It's called "OCR B" and is a world standard machine and human readable font.  It is used in perhaps hundreds of thousands of different applications worldwide since its development in 1968.

As for the collectibility of the 1979 series, the argument of 7-digit radars not being true radars is pretty much on the same plane as saying that modern insert notes are not really replacement notes because they look like regular notes.

I agree that 7 digit 1979 radars are collectible because I have one:

52440544450

I found it at my job way back in 2004!   

Here's the link to that thread.

http://www.cdnpapermoney.com/forum/index.php?topic=967.0

Dean