Pioneer Serial Number Analysis - early returns

[Dr. Strangelove]

For almost a year now, we have been collecting serial numbers. We know about the first two letters and how they indicate the Year and Month of manufacture.

One of the goals was to collect enough numbers to help identify a pattern which would indicate manufacturing counts of each component model that Pioneer produced. Thus far, this goal has not been achieved.

For example:

WB-3902676M SX-1250
WC-3902679M SPEC-1

The W means 1976. The B is February, the C is March. These are two entirely different components, yet, numerically, the serial number differs by 3.



OK, so let’s pick two components with the same year and month:

VK-3903320M TX-9500
VK-3903439M TX-7500

Here, these tuners were both made in 1975 (V), November (K). They differ by only 119. Yet, they are two different tuners.



See what I mean? It gets worse.

VL-3400096 RT-1011L
VL-3949496 SX-737

Here, both of these were made in 1975 (V), December (L). Two completely different components. Yet, they differ by 549,400. Hard to imagine that Pioneer made more than half a million components in a month.



And, many more confusing examples can be found.

I suppose we can come up with all sorts of ideas as to why such patterns exist, or do not exist. Maybe certain blocks of numbers were reserved for certain components. Maybe one manufacturing plant had a block of numbers, different from another plant. And, remember that some numbers were stamped onto the back of a component; others were on aluminum tags, while still others were on paper tags. Some aluminum tags were riveted, some had plastic expander pins, some had screws. No consistency can be found here.

In the end, many, many more numbers are needed to identify a pattern, assuming one exists in the first place.

I would love to be able to determine manufacturing counts by analysis of serial numbers. However, not enough data exists at this point.

The only thing that is consistent (and there are exceptions in this) is that Pioneer stayed with their YEAR and MONTH convention from 1954 to at least the late 1990s.

If you have not yet added your serial numbers, please do so in this thread:
http://audiokarma.org/forums/showthread.php?t=60932

Thanks,
Doc

 

[Mark W.]

Doc if I can add something to your thinking using Canon FD mount cameras as an example.

Canon uses a VERY similar numbering system only they seperate the date from the serial number using a stamped ink number to denote the manufacture date and the factory a piece was made in.

Their serial numbers are all consecutive and all start from the same point.

BUT EACH thing be it Camera body, Lens, Viewfinder, Flash what have you has it's OWN serial number run.

So in your examples if we use this same method above comparing two different units is of no value at all.

You need to compare the exact same unit with others just alike.

SO if you have four numbers for say a SX-1010 and they are sperated by 5,000 low to high then you could say that between this month and that mont Pioneer made at least 5000 units of the SX-1010. This number will always be changing until you find the lowest and highest number.


Now using Canon Lenses as an example they ALL have their serial number runs start at 10,000 so the first lens of that paticular model would be number 10,001 And the last would be the highest number what ever that maybe.

I own a Canon 85mm f2.8 nFD Soft Focus Portrait lens these are an extremely rare lens to find. Over the course of a year maybe 2 or three would show up on ebay and maybe a couple more might be in used dealers stock. We started to collect as many numbers from as many sources as we could find even to the point of calling dealers to get numbers off lenses in their inventory and using the three largest Canon FD forums to find those known in collectors hands.

We then put them all together and using the dates of manufacture as well have determined that not more then about 2500-2600 of these lenses were made. WE were lucky in that this was a lens made very very late in the systems run and we got a couple numbers with dates very close to the end of all Canon FD production giving some assurance that the high numbers we had collected would not be eclipsed. And then using the average of how many were made over the lenght of manufacture we were able to get pretty accurate with only having a few dozen numbers to base our theorys on.

Hope this helps

 

[Dr. Strangelove]

Yes, I have looked at it the way you described, and there is just not enough data collected at this point. And, using your method, there are still anomalies.

For example,

TX-7500 ZC-3400171M (1979 March)
TX-7500 VK-3903439M (1975 November)

Maybe another 200 or so numbers will help.

Doc

 

[Mark W.]

Yes, I have looked at it the way you described, and there is just not enough data collected at this point. And, using your method, there are still anomalies.

For example,

TX-7500 ZC-3400171M (1979 March)
TX-7500 VK-3903439M (1975 November)

Maybe another 200 or so numbers will help.

Doc

Well yes I would say that does cause some problems but maybe not the one you think.

If you check your Pioneer data base for the TX-7500 tuner it lists the dates of manufacture from 1975V to 1978Y So unless my math is wrong how can you reconcil these two sets of numbers if your first example is shown to be manufactured 3 months after the end of the run.

one or the other number is incorrect. I would lean towards the manufacture date given in your above example being the one off namely the year. Unless this date code could be verified I would tend to think it was in fact a unit made in March of 1975 very near the start of production as the numerical sequince would then make sense.

March #171 in the run being made

November # 3439 being made

giving a monthly run of 382 (first of Mar to last day of Nov) to 466 units (last day of Mar to first day of Nov) being made.


Again only proposing possiblities based on a tiny amount of info available for this discussion.