Gramophone Co. HMV... How many copies?

[Inigo]

I'm wondering how many copies of each Gramophone Co record were sold or made until they changed the letter in the runout area in the series G R A M O P H L T D...?
I've been trying to calculate, but there are data I'm missing...
From one single matrix (wax) no1 they took a negative (mother) no2. In the early times they used this no2 for pressing records directly. It wore out quickly, for each one could press from 500 to 1000 78s prior to death.
Later (when?) they introduced another step. From no2 they took positives no3 (how many?) and from there, the pressing metal parts no 4 (how many?) from which they pressed records. This multiplied the records that could be pressed.
At which step they introduced the master letters GRAMOPHLTD?
Yet later, another step more (no5 and metal pressing parts no6) so it multiplied by infinite...
This makes me crazy.... It's there any way to guess how many copies with the same mother letter GRAMOPHLTD of a certain era exist?
If someone can add some light to my question, thanks in advance!
I know it's difficult, because many records of the old ages (opera, for instance) were reprocessed for reissue... what makes the accounting yet more difficult...
I suppose one has first to try dating the pressing one has under examination, for knowing what manufacturing process was in use then (i.e. how many steps and when the master letter was introduced, etc....)

[recordmaker]

full the process is wax, master mould, mother mold, stamper the master mould was also initially used as the stamper with the limitation of 500 to 100 copies per take.

The problem was that the earlier methods of making the mother mould were all processes similar to pressing the a record in a plastic material and then re-graphing it and making more stamper electrotypes from it. These moulding materials and the additional graphing process all lead to a very variable quality of stamper production in the early period (1900-1910ish I think)
This also creates a need a need for more mother molds than used in the later all metal process as the material is inclined to deteriorate in the plating process each time.

A metal mother mould could create stampers until it was damaged in the process which up was probably due to accidental damage rather than wear.

However in the multi-national Victor/Gramco/DG business I assume that shared metalwork was distributed as metal mothers to each company to derive there own local stampers, master moulds from the earlier period would have been used to make a metal mother mould

[Inigo]

Thanks. And these mothers, from which stampers could be made, are the ones carrying the GRAMOPHLTD letters? Or they are the later stampers who were assigned letters? (Earlier these were Roman numerals). In any case, the highest I've seen is GG or GR, so the eleventh, then. In Roman numerals I think I've heard of some XXIV, 24th one, being common to find numbers in the teens.
It seems not much... depending on how many stampers could be made from each mother.
Maybe I'm writing and the ones bearing the numbers were already stampers.... Still we don't know how many stampers could be derived from one mother. Could it be 10 or so?

[shoshani]

However in the multi-national Victor/Gramco/DG business I assume that shared metalwork was distributed as metal mothers to each company to derive there own local stampers, master moulds from the earlier period would have been used to make a metal mother mould

This is generally correct, although there are exceptions. In Victor pressings of Gramco/DG records, they often retained the original matrix in the dead wax. When they didn't, they impressed a crown in the dead wax, but didn't dub the record. Likewise, I have plenty of HMV records derived from Victor and Bluebird material, all of which carries the original US catalogue number in the dead wax from the mother moulds.

[recordmaker]

I am more acquainted with the manufacturing processes used than the study of matrix and other markings on the discs the following link shows a useful summary of these letter markings.
http://early78s.uk/markings-on-78-rpm-records/

The later ( possibly 1910 on) metal mothers could be used to make as many stampers as needed however the limit of one stamper per day per mother at he time would mean that a popular record may need more than one mother to produce enough stampers in a short time.
it is interesting to note that as far as I can see the mothers are not numbered however damage tot eh mother would not produce a working stamper so would be rejected at an early stage ( the first test record of the press) whereas stampers can get scratched or dented in use and need to be removed from the press as soon as possible.

[Inigo]

Thanks for the info; good friend Norman Field made this website, I had forgotten about it! He says the numbers were applied to the stampers, not mothers as I believed. So maybe, irrespectively of which mother had produced the stamper, they were all numbered sequentially.
So that goes the desired guess... If we see a 1914 record with a big G, it means first stamper, so there must be between 500 and 1000 copies like it, etc. etc. With roman numerals, if the first stamper carried no number, then the I, II, III actually meand 2nd, 3rd, 4th stamper and so and so...
Thanks all for your help in clarifying a bit this question!

[recordmaker]

Although it is reasonable to estimate 500 to 1000 copies from stamper it is possible for a stamper to be damaged after only a few copies have been made so the G stamper could have made only a 100 records.
The pre 1918 manual lose plate pressing allows for about 20 records per hour per stamper (160 per working shift ) 360 records as they handled 2 moulds at a time, assembling one while another was pressed.
So a working stamper should hold up for a week or more however the lose plate method is more prone to accidental damage in my opinion than one in a fixed press.
Around 20 % of records pressed were rejected for faults and some percentage get damaged in transit to the customer.

[Inigo]

That would mean 300 copies x 6 days = 1800 copies from two stampers, so the mean between 500 and 1000 it's a fairly good estimate. Only half of them carried the same letter (two stampers, two letters). So if I have a record with a G, it means there were only 250 to 500 exact copies with the same letter. In one country or factory alone. Let us think that the original plant shared 10-20 mothers with other factories. This makes a total of 2500 to 10000 copies from different factories carrying the same letter (with different labels).
Still it seems a miracle that one copy has survived, and yet more difficult that one record such as Il balen del tuo sorriso by Titta Ruffo can be found in so many places today! I have owned two copies, and I'm not a collector who buys lots of records! Let apart certain Caruso records that seem to be in every one of ten record piles! I'm talking about the typical pink label one side Gramophone record!
Much more copies must have been made.... Still it's kind of a mistery...