Tech Musings

This section of the Richards Laboratories website is an eclectic grouping of highly esoteric bits pertaining to the overall methods and philosophy used to produce these blanks. It is intended for those who are interested in the details.

MR. RICHARDS AT THE CLOSE OF 31 YEARS OF THOUGHT AND RESEARCH, THEN 5 MORE YEARS OF WORK MAKING BROWN WAX CYLINDER RECORDING BLANKS.

---AUGUST 16, 2013

This is the longest continuous session of labor he ever performed

William Shakespeare (1564-1616)

from Macbeth, Act IV, Scene 1: A dark Cave. In the middle, a Caldron boiling. Thunder.

Enter the three Witches.

First Witch
Thrice the brinded cat hath mew'd.

Second Witch
Thrice and once the hedge-pig whined.

Third Witch
Harpier cries "'Tis time, 'tis time."

First Witch
Round about the cauldron go;
In the poison'd entrails throw.
Toad, that under cold stone
Days and nights has thirty-one
Swelter'd venom sleeping got,
Boil thou first i' the charmed pot.

ALL
Double, double toil and trouble;
Fire burn, and cauldron bubble.

Second Witch
Fillet of a fenny snake,
In the cauldron boil and bake;
Eye of newt and toe of frog,
Wool of bat and tongue of dog,
Adder's fork and blind-worm's sting,
Lizard's leg and howlet's wing,
For a charm of powerful trouble,
Like a hell-broth boil and bubble.

ALL
Double, double toil and trouble;
Fire burn and cauldron bubble.

Third Witch
Scale of dragon, tooth of wolf,
Witches' mummy, maw and gulf
Of the ravin'd salt-sea shark,
Root of hemlock digg'd i' the dark,
Liver of blaspheming Jew,
Gall of goat, and slips of yew
Sliver'd in the moon's eclipse,
Nose of Turk and Tartar's lips,
Finger of birth-strangled babe
Ditch-deliver'd by a drab,
Make the gruel thick and slab.
Add thereto a tiger's chaudron,
For the ingredients of our cauldron.

ALL
Double, double toil and trouble;
Fire burn and cauldron bubble.

Second Witch
Cool it with a baboon's blood,
Then the charm is firm and good.

Detailed batch summary 

Some definitions:
Stearic acid is known as a "fatty acid". C18H36O₂ (called also just "stearic" in industry)
Saponification: Conversion to soap.
Saponify: To change a fat or an oil into soap by treating with an alkali. Saponification is the hydrolysis of any ester into the corresponding alcohol and acid. Hence, any hydrolysis.
Soap: Normally thought of as a cleansing agent. Soap is the end result of the reaction between a fatty substance and an alkali. It is the result of that acid/base reaction. This particular variety of soap used to produce the brown wax contains both sodium and aluminum metals. It also does not dissolve in water.

Edison said of it: "It's a soap that doesn't lather".

282.4 grams stearic acid (total)
211.8 grams of stearic acid is saponified using 10.75 fluid ounces of hydrated aluminum solution.
The remaining 70.6 grams of stearic acid is added after the initial 211.8 grams has been saponified.

Hydrated Aluminum Solution:

16 fluid ounces distilled water.
32.4 grams of lye (sodium hydroxide NaOh)
1.07 grams of aluminum metal (6 inch square x .0005 inch thick foil)

Method:

Lye is added to water and stirred until completely dissolved. This creates heat. Solution is allowed to cool a bit before adding the aluminum. The aluminum foil square is torn into 3 strips each 2 inches wide and those are rolled up into marble-sized balls.

After the lye/water solution has cooled a bit, the first foil ball is dropped in. As the aluminum dissolves in the lye/water, more heat is created, and more cooling time is required. The foil is added in 3 separate balls in order to keep the overall amount of heat down low enough not to crack the glass. Several minutes are allowed between adding each foil ball. After the 3rd foil ball has been added and after the final solution has cooled down, then it is filtered using 2 coffee filters together. The solution gets filtered that way 3 times. Then it is stored in a plastic jar overnight to settle. The following video shows the aluminum dissolving in the lye/water solution:

The next day, it gets filtered twice more. A total of 10 coffee filters is used for these 5 times of double-filtering. The solution should be water-clear at the time of being used to make brown wax.

The 10.75 fluid ounces of 5X double-filtered hydrated aluminum solution which gets added in 16 portions to the 211.8 grams of melted stearic acid saponifies the stearic acid. The resulting compound contains both sodium stearate and aluminum stearate. The aluminum helps to make the blank shiny and smooth and when the amount of aluminum is just right, the noise level has a distinct low spot in the graph which shows aluminum amount plotted on the "X" axis, and overall background noise level plotted on the "Y" axis. Either an excess or a deficiency of aluminum or saponification can result in more surface noise than ideal. Each different brand of stearic and each different base used to make the stearic has a very slightly different balance of these factors in order to arrive at the lowest possible surface noise. Stearics are available in different bases such as animal tallow, palm, and various vegetable bases. Each has its own balance requirements when making brown wax from it.

Back 120 years ago when brown wax blanks were modern technology, the only type of stearic available then was made from animal tallow. Richards Laboratories uses an available modern stearic which is made from animal tallow.

After completion of the saponification, the remaining 25% of the stearic acid is added and thoroughly cooked in. This increases the capillarity, and decreases the viscosity of the whole. It becomes a very runny, watery consistency, very thin and very liquid. To this saponified and thinned stearic is then added 59.1 grams of ceresin wax. The ceresin is allowed to cook in for 10 minutes at 232 degrees C.

The ceresin is tempering. It softens the wax. The amount of ceresin can be varied to obtain harder or softer recording blanks. A good "hot day" blank might have around 17.3% ceresin, while a better "cool day" blank might contain 18.5% ceresin. At a usual normal room temperature somewhere between 72 and 75 degrees, either of those values will work fine, the 17.3% has a bit less noise than the 18.5%.

The ceresin also serves to make the metallic soap compound non-hygroscopic. That means it does not absorb moisture.

The yield is 334 grams of brown wax. At first this might seem rather odd, because if the weights of all the solids are added up, the result is 282.4 grams stearic + (32.4) x (10.75/16)=21.77 grams lye + (1.07) x (10.75/16) = .72 grams aluminum + 59.1 grams ceresin wax totals up to 364 grams. But the total yield is 334 grams.

30 grams is lost to vapor in this open process.

Each cylinder casting uses 120.3 grams of brown wax.

Disclaimer: This process is very dangerous. It creates hazardous fumes, and there are high temperatures involved. Risk of serious injury is ever-present, and all of the above information is provided for explanation purposes only.

The following video is a brief overview of the ingredients and process for making the brown wax:

Scrap Run Blanks

A typical wax batch in the present process can cast 2 blanks. After casting those two blanks, there's enough wax left over for about another half of a blank. That leftover wax is then weighed and saved.

Same for the cutoff ends of the cylinder castings. That leftover wax plus the cutoff ends are scrap wax which can be used to make more blanks. There are many references in Edison's original process which tell of adding a small amount of scrap wax into a fresh batch of wax. 

So far at least, I have not been adding scrap into fresh batches because the fresh wax just works a little too nicely to tinker with any farther. Sure, it can most likely be done, but the present process flow uses the fresh wax alone for the first two castings without adding any scrap. This is being written in December, 2014 and it documents this time period.

Up until close to the end of the 2014 blank production season, I was casting the 2 blanks from the fresh wax, then immediately after those two blanks were finished, shaved, and tested, I'd move right off into cooking up a small scrap batch just enough to cast one more cylinder.

That mixing and cooking is fairly involved and takes the better part of an afternoon, just to get that little scrap batch ready for casting the 3rd blank. It involves scraping all surfaces of the (2) cutoff ends from the previous castings to clean them up making them ready to melt and reuse. Those (2) cutoff ends, plus the (1) cutoff end from the previous Scrap Run blank, plus the leftover wax from the recent fresh batch add up to enough wax to cast another blank, and then to have some amount of leftover scrap wax.

After having used this general process format for several years it slowly became evident that a small amount of fresh stearic needs to be added to these scrap batches in order to keep the wax properly conditioned for casting. Instead of waiting until the need for additional stearic shows up as a visible sign in the castings, it was found that there is a small standard maintenance dose of stearic which can be added to every scrap batch which keeps these batches under constant control, and as such, no visible signs of a stearic deficiency ever show up 

In rough terms, it works like this: If you wait until problems show up such as excessive shrinkage, then about 2.5% by weight of added stearic is required in order to bring the batch back into balance. The price one pays for proper conditioning for good casting is that any too much stearic starts to bring up the noise level slightly. More stearic means more noise. Not enough stearic brings on surface wrinkling and excessive shrinkage. It is a very touchy and sensitive balance which must always be watched because it changes every time the wax is reheated to be cast.

On the other hand, if a smaller amount of fresh stearic down around in the range of 1.0% to 1.4% is added to each small scrap batch every time, then the noise can be kept very low and also the casting properties are maintained and excessive shrinkage  and wrinkling are avoided.

But that's all a lot of work to go through each time in order to cook up a perfectly balanced scrap batch to cast only one blank!

A more efficient method is this:
Fresh wax batch: Cast 2 blanks.
Save leftover wax.
Fresh wax batch: Cast 2 more blanks.
Save leftover wax.

Now take the (4) cutoff ends plus the leftover wax from the (2) fresh batches, plus the cutoff (2) ends from the previous scrap run blanks, weigh that all up and add about 1.25% stearic and cook that up to make it ready for casting. This now yields (2) more scrap blanks, with plenty of leftover scrap for next time.

The end result of this extension of the process is that now with this method, every cooking session yields 2 blanks. It makes better use of the time involved in cleaning the scrap, weighing it, then calculating and weighing the added stearic plus it better makes use of the cooking time.

Meta Scrap

Meta scrap is a term I came up with which attempts to relate the phenomenon of an endless scrap cycle which contains an ever decreasing amount of scrap from the very beginning of a scrap run. The leftover scrap is constantly being renewed with recent scrap each time it is used, however some of the original scrap is always present although in a progressively decreasing amount.

Example: Say we start out with a fresh wax batch and cast some blanks, saving the leftover wax plus the cutoff ends. For this example we will assume there is no prior scrap saved, this is the start of a new scrap run. As such, the only scrap now available are the two cutoff casting ends plus the leftover fresh wax. Available scrap: 1 tail end of fresh batch, plus 2 casting ends. Not enough to make more blanks.

Another fresh batch is made and 2 more blanks cast. This now yields 2 more ends plus 1 more piece of leftover wax from this batch. Now the total available scrap is: (2) leftover chunks from (2) batches, plus (4) cutoff ends.

This scrap is cleaned, combined, stearic added, then cooked. It is used to cast 1 more blank which is a scrap run "SR" blank. (These blanks are noted on the title end with the letters "SR" in the serial number extension to indicate they are Scrap Run blanks)

The whole gist of the Meta Scrap idea now comes along: Consider the leftover scrap wax remaining after this (1) "SR" blank has been cast. It is leftover scrap.

This leftover scrap is not quite enough to cast a blank so it gets cleaned, weighed, and bagged.

Now another fresh batch is made and 2 blanks cast from it. A 2nd fresh batch is made and 2 more blanks cast from it.

Total available scrap now is: (2) leftover chunks from ends of those 2 recent batches, plus (4) cutoff ends from those (4) blanks, plus (1) cutoff end from the prevous "SR" blank plus the leftover bagged Meta Scrap from the previous run.

Now the process starts to sink in and take hold because this Meta Scrap has wonderful audio and casting properties as it ages throughout many repeated cookings.

Soon the amount of meta scrap builds up so that it can't always be all used, because it needs to remain roughly 40% of the total in any one "SR" batch. So, arriving at proportions of recent scrap being 58.75%, meta scrap being 40%, and fresh stearic being 1.25% becomes a requirement when weighing out the ingredients prior to combining and cooking.

The gist of this is that the Meta Scrap chunk always gets added to the most recent scrap batch. This can continue for years until the Meta Scrap starts to get so contaminated that it must be eventually set aside. When this happens, a new scrap cycle is started and it takes a few of the smaller sub-cycles in order to get it up and running to its fullest working mode.

A typical long scrap run can last for a few years and extend over the course of 100 blanks or so before the Meta Scrap needs to be set aside due to contamination. This contamination can be envisioned as being roughly the same as boiling water repeatedly in a tea kettle. Eventually the minerals build up because the water evaporates, leaving them there. More water is constantly being added and boiled off but the minerals do not boil off, they instead stay in the kettle where they accumulate.

Steam locomotive boilers work in exactly this same way, they eventualy concentrate minerals which must be removed by unscrewing the washout plugs and washing out the boiler to get the gunk out.

Brown wax for cylinders works this way too because the stearic is volatile, and it does boil off and it escapes as vapor, leaving solids behind which stay there. Only with the cylinder wax, there is very little of this activity, the accumulated contamination is very small indeed, so it takes years for it to build up to an intolerable level which could lead to objectionable noise in high quality blanks.

The contaminated meta scrap is saved for future experiments where by some sort of micron size filtering can very possibly be used to remove the contamination. It is excellent wax, it just has a bit too much gunk in it for first-rate cylinders.

Serial Number Coding 

Now that some background on this topic has been explained, here are the serial number codes which document all of it:

The blanks which are made from the fresh wax with no scrap added are just the plain numbers such as M267.

The Meta Scrap blanks are noted as follows:

M268-SR1B19. Those numbers mean:

Blank M268, Scrap Run number 1, blank number 19 of Scrap Run 1.

It should be noted that as a Meta Scrap run progresses, it still contains some few molecules of the first batch of wax which was used to start the run. The amount of this original wax used at the start of the run is always ever-decreasing, but it never gets all the way to zero. It approaches zero, but it never gets there.

Kind of the same as getting ever closer to something by halving the distance from it each step you take. You will get close, but you will never arrive at that location.

This has been an explanation of the "SR" series of dark blanks.

A brief look at the math behind it

Here are some actual numbers to add a bit more detail to the above verbal description:

Let's start with a typical 334 gram wax batch. Two castings are made, each of which weighs about 120.3 grams. That leaves 93.4 grams of wax left over after having cast those 2 blanks.

Do this three times and now there is about 280 grams of wax left over from the 3 batches. Each cutoff end from a casting weighs about 13.37 grams. There will now be 6 of these ends, so that totals about 80 grams. Add it all together and it comes to about 360 grams of available scrap wax. (in actual practice this can vary between about 330 ands 360 grams)

To prepare this mixture for casting, a small amount of fresh stearic is added. The range can be from around 1.25% to 1.40% depending upon the overall condition of the wax. If 1.4% is used that would come to adding 5 grams of stearic. The whole is heated to 232 degrees C and held there and stirred for 20 minutes, then allowed at least 5.5 hours to completely cool before casting.

Usually about 3.4 grams will be lost to vapor from this cooking.

Two Scrap Run blanks can be cast from this batch.

That will then leave about 120 grams or slightly less left over. This is the start of a new scrap run. This is the first new meta-scrap. It gets weighed and bagged, along with the cleaned-up ends of the 2 scrap run blanks just cast. That total amount of saved meta-scrap will be about 147 grams.

Next, 2 more fresh wax batches are made and 4 blanks cast. This leaves 93.4 grams times 2 of leftover wax. (187 grams), plus the 4 ends being 13.37 each for a total of 53.5 grams, and that all adds up to 240.5 grams.

Add this all together and the total scrap becomes 387.5 grams. Note that 240.5 grams of this is the "recent" scrap from the 4 blanks just cast, and the remaining 147 grams is the "meta scrap" left over from the previous scrap blanks cast.

Note also that 147/147+240.5 = .3794 which shows that this scrap mixture contains 37.94 % meta scrap in the total amount of 387.5 grams of scrap.

The target is 40% meta scrap, and it takes a few cycles to build up to that before meta scrap has to be left out in order to hold the amount down to 40%.

So now that all gets put into the pot and weighed, then 1.4 % stearic added (387.5 x .014 = 5.4 grams). Then the whole gets heated to 232 C, cooked for 20 minutes and the 2 blanks are cast from it.

This process gets repeated many times, always re-using the leftover meta scrap and the recent scrap. It slowly converges on the 40% mark, then it needs to be held there by only using enough meta-scrap each time to assure 40% meta scrap.

So, over the long term, it very closely averages out to 40% meta scrap.

In order to be able to calculate the remaining amount of original scrap contained in any "SR" blank as this continuous process plays out, it is easily solved by calculating using .4 as the factor.

The first SR blank contains .4 scrap by weight. The 2nd SR blank contains .4 times .4 original scrap by weight. The 3rd SR blank contains .4 x .4 x .4 original scrap be weight. The content of original seed-scrap remaining in any successive SR blank can be calculated by taking the number of the blank in the run and multiplying it by .4 raised to the power which is the number of the scrap blank within the particular run it comes from.

Here is a table of the first few:

It approaches zero, but it never quite gets there.

And now for something completely different.....