EH Scott Radio Enthusiasts

The Fine Things are Always Hand Made

As most of you know, I've been gathering and analyzing Scott serial numbers for over 20 years. I have a couple requests....first, if you haven't sent me your serial numbers, please do. Also...if any of you have a confirmed date for a set/serial number, please send me that. There is a warranty tag on ebay at the moment, these are great, since they have the date and serial number on them. A letter with a date is also good confirmation. I would greatly appreciate any info you fellows can provide. I'll post more about serial numbers here in the future.


Views: 15881

Reply to This

Replies to This Discussion

Survival rates are not a reliable measure of production.  Survival rates vary greatly across brands and even across models within a single brand.  The serial number analysis avoids the ambiguity of survival rate.


I agree survival rates have a lot of variables and that means more room for error and I could be way off here.  One thing is for sure is that Zenith published their 1930s chassis production numbers in the front of their #2 service manual. Since the 1000 Z is the only Zenith model using the 2501 chassis you have a sold production number to project off. The average Scott survival rate, less the AW 27 estimate Norm posted, you get a 7.4% survival rate across all Scott models listed. (See updated chart below) Norm's AW 27 numbers are constant with the AW 23 and Philharmonic percentages so his number for estimated AW 27s produced make since when you look at them in this context.  Did Scott ever publish their production numbers anywhere or are these estimated productions numbers listed here a guess based on a higher low range of collected serial numbers? 



AW23              551              5650           9.8%

AW12              239              4675           5.1%

Phantom         317              3900           8.1%

AW15              264              3725           7.1%

Philharmonic  378              3730         10.0%

16/18             109              1530           7.0%

Laureate           41                900           4.5%

Masterpiece     58                 740           7.8%

Super12           52                 720           7.2%

Ave. Scott survival rate                        7.4%


AW 27              25                 275           9.1%

1000 Z             53                 350          15.1%

Any record of Scott production numbers is lost.  However here's something to think about.  All possible combinations of a letter plus up to three digits and two equal letters plus up to three digits yields a total of 52,000 possible unique serial numbers.  No Scott serial numbers have been found with a single digit leaving a total number of 51,480 serial numbers.  Scott allocated one serial number to tuner chassis and a second serial number to power amplifiers hence a possible 25,740 allocated to tuner chassis and representative of the total production of Scott sets with serial numbers before WWII.   Only the serial numbers on tuner chassis have been collected and analyzed.  As previously mentioned (somewhere) I have seen one late Phantom with a three equal letter, two digit serial number indicating that the one letter and two letter serial numbers were substantially fully allocated.  By comparison Kent's serial number block analysis has yielded a total production of 25,570 sets, an insignificant difference from the total possible number of sets based on full allocation of of single and double letter serial numbers.  It is safe to state that Scott's total production of serial number sets up to WWII was 25,500 to 25,800 and that Kent's estimates of model number production are robust.


Wasn't there some discussion a number of years ago about searching patent rights contracts and payments to the patent rights holders? Didn't Scott have to report how many chassis he made that used the various patents? I thought he had to keep track and pay the contracted fee. Thus RCA had some idea how many radios Scott was selling.

It's too bad Scott's engineers, Clay and Hobbs are not alive today for this discussion. I think this is good direction to focus on if we are to verify Scott's actual production numbers and you would think that RCA, like Zenith would have keep these kinds of records. Zenith archived examples of what they produced, much of their internal company records. All this information still exist in Chicago, even under LG's ownership of the Zenith name and licenses, today.  Maybe RCA's records also exist somewhere. One trend that does seem to appear in Kent's serial records is that the better radios, like the AW 23, AW 27 and Philharmonic, seem to have a slightly better than the Scott average survival rate. This leads one to speculate that the high quality, high-fidelity, wide band radios in fancier cabinets were kept around longer because of their looks and high quality reception characteristics. 

Auroraoldradios said:

Wasn't there some discussion a number of years ago about searching patent rights contracts and payments to the patent rights holders? Didn't Scott have to report how many chassis he made that used the various patents? I thought he had to keep track and pay the contracted fee. Thus RCA had some idea how many radios Scott was selling.

Kent's serial number analysis reflects two expected survival rate trends.  The survival rate is not only lower for lesser models, it is lower for earlier models.



You are saying that in the database there are sets scattered throughout the approx 52,000 combinations as one would expect if they used up all combinations?  As I stated previously, I only have the Philharmonic numbers and there seems to be no rhyme or reason to the numbers.  They are scattered in blocks.   Seeing my numbers it appears that they just randomly picked an opening in the serial number sequence and started there.  I guess the serial numbers would make more sense seeing the whole database.

Regrettably, amp serial numbers were not collected along with what variation of the set belongs to that number.  

Apparently blocks of serial numbers were selected somewhat randomly across letters but not so much across number lots (some order to numerical sequence) and serial numbers were likely assigned within letters until the block (letter or letter letter plus a series of 200 or so numbers) was exhausted.  Serial numbers from multiple letters might have been concurrently assigned.  The AWA Review, Volume 19 presents Kent's early serial number block analysis which may clarify the issue somewhat.

Norman provide more precise information, here is more detailed information regarding the prefixes and blocks for the Philharmonic (since Scott is most interested in them):

D 650-850

E 650-850

F 650-850

AA 10-250

BB 10-500 (entire prefix)

CC 450-500 (odd block behind Phantoms)

GG 250-500

KK 10-250

LL 10-500 (entire prefix)

RR 10-500 (entire prefix)

XX 10-150 (odd block before Phantoms)

YY 250-500

If we add up these blocks of serial numbers, including the 3 early prefixes, we get a production volume of 3240.

All records of the company were lost in the fall of 1957. The production records were moved from Chicago to Plymouth, IN in the early 50s when they relocated out of the city. Following the bankruptcy auction, items not purchased were hauled to the landfill there in Plymouth. I have two separate accounts of this, one person remembered many filing cabinets being hauled to the dump. Many of the dies and molds (think escutcheons, etc.) were also dumped.

RCA records of sales have not been found...I've corresponded with a couple of notable RCA collectors, but they haven't seen anything like the production reports for any company other than RCA. I suspect those were reported to legal/accounting and never kept like line production records.

A few of the late, double-letter prefixes are chopped out outside the 10-250 and 250-500 ranges. These include:

HH (FM Tuners)
JJ (Specials)
NN (Masterpiece)
UU (Laureate)

Even within these, there are boundaries, sometimes in groups of 100 or 150. The block for Specials in JJ is actually a block of 50...but production numbers give us no more than 21 sets.

I have more precise data based on known serial numbers for all models. This analysis can be done for every model. I don't share all the detailed info - I have spent years gathering it and still have plans to do further analysis. I've had several data experts review the data too - the conclusions published to date are validated by data analysts. I honestly try to err on the side of "we don't know"....


To the discussion on survival %, I will give this:

Set              Estimated Known    %SurvivalSet 

AW23          5307        551          10.38

AW12          3963        239             6.03

Phantom      3457        317            9.17

AW15           3322        264            7.95

Philharmonic 2970       378            12.73

16/18          1396        109            7.81

Laureate         883          41            4.64

Super12         642          52            8.10

Masterpiece    512         58             11.33

FM Tuner        132           9             6.82

Special              21         10            47.6

The average survival rate is 8.18% - call it 8% for general discussion. This is based on <known serial numbers>, not the "blocks" defined in my previous post...Obviously, the numbers would change for the block production estimates..

One factor I think contributes to survival is cost...the higher priced sets seem to do better, if you paid that much for a radio, would you throw it out?'

I use the 8% number as the average, although the block calculation is closer to 7%...


Right Kent.  I have the full Philharmonic data set, but the problem is with the amps.  I have a KK-118 amp sitting right in front of me.  Other amps I have are RR-282, GG-436 and KK-100 among others.  These Philharmonic amp serial numbers all fall within current tuner serial number blocks, so the question becomes, if the amps had unique numbers in the Philharmonic blocks,  wouldn't there be half as many radios produced?  Has anyone seen an amp number that was the same as a tuner number?

Short answer to your question - I have several examples of amp numbers the same as tuners. However, no examples where the tuner and amp matched. I suspect that perhaps the amps got the same number in grease pencil as the tuner at the start of build, but when shipped, they might have been with other tuners. Of course, there are differences between the sets/amps and some should/needed to stay together. Also the early amps didn't have numbers like the later amps. But yes, the numbers do overlap.


Reply to Discussion


© 2024   Created by Kent King.   Powered by

Badges  |  Report an Issue  |  Terms of Service