MAKING BROWN WAX -- THE BATCH SHEET, HOW IT WORKS and HOW TO USE IT
Click Here to see the blank and filled out batch sheet.
The use of this standard batch sheet came about because before I used it, I had to copy down the columns and target data each time by hand in the lab notebook. As you might guess, copying the blank form by hand each time before making a wax batch gets rather tedious.
In the beginning, as things were still being fine-tuned and adjusted, the handwritten individual forms were appropriate because each batch differed very slightly. However, after having centered in on this particular process and after having used it many times over several years, then the standard form became appropriate.
Back at the very beginning, the batches were "free running" in that the time allowed was open-ended. In other words the total time for each batch varied somewhat, being dependent upon taking all the time needed to successfully cook in the precipitate produced from each of the hydrated aluminum additions. Back then , even the total number of these additons was not standardized. Some early batches used only (12) or (14) additions, and some batches used as many as (20). These batches have always been about the same size, they produce two cylinder castings with enough leftover wax to make about another half of a cylinder. The total amount of hydrated aluminum added to this batch has always ranged from around 10 to 12 fl oz. Total time for these early batches ranged anywhere from around 2 hours to sometimes close to 3 hours. It is now standardized at 2 hours, and 26 minutes, always using 16 additions of hydrated aluminum measured per the schedule starting with 10.75 fl oz.
Larger batches were attempted a few times, and it was found that many other factors creep in when doing this. The overall quality of the resulting wax was low, the blanks were noisy, so I simply went back to the smaller batch size which I know works properly.
The total amount of hydrated aluminum added always has been roughly the same. Certain brands of stearic work well with 11 fluid ounces of hydrated aluminum for this batch size. The particular brand and type of stearic being used currently as of 2016 works best with 10.75 fluid ounces of hydrated aluminum solution, added in 16 portions. The type of lye used also can slightly change the overall amount of hydrated aluminum liquid added.
The batch sheet uses lots of shorthand abbreviations for things which go back to some of the original tests and methods used along the way of learning. Rather than change the terminology, I have found it easier to just keep the original nomenclature intact.
I will do my best in the following explanation to clarify each one of these shorthand references. Note: The following is a very long, involved, convoluted explanation. It has been written in an attempt to make the use of this batch sheet available to all.
For those who wish only a short and rough idea of it: It can all be summed up this way: The batch sheet works as a dynamic guide to making brown wax. It shows when the batch is on schedule and it indicates what to do if or when it drifts off of schedule. When followed as closely as possible, this schedule provides excellent uniformity from one batch to the next.
WEIGHT UNITS
As for the weight units used, each unit is equal to .2674 grams. So, when ever you see a weight specified in "washers", just mulitply that by .2674 to get the weight in grams.
SPLASH
The term "Splash" refers to the addition of a small amount of hydrated aluminum solution to the melted stearic acid. There are (16) such additions which take place over a time of about 1.5 hours. Each such addition is numbered in the "Splash" column on the batch sheet.
AMOUNT OF EACH ADDITION
To understand this clearly takes a bit of translation. It is a simple enough concept to understand, but the way it's presented here can be very confusing. It's confusing due to the method used. The method makes things easy, simple, accurate, and fast while doing it, but it also makes explanation somewhat convoluted.
The first 10 additions are clearly labeled in fluid ounces in the column which has "10.75 fl. oz. start" at the top.
Due to the tapered shape of the bottom of the plastic jar used to measure the hydrated aluminum, after addition #10, the units change to inches. The place on the jar where the shape changes from straight cylindrical, to tapered is called "Top of Bevel".
In order to be useful elsewhere for this method to be exactly duplicated in any other place, at any time, additions 11 through 16 must be translated to fluid ounces so that all (16) additions are specified in fluid ounces.
Here are additions 11 through 16 in fluid ounces:
Addition | fluid ounces |
#11 | 2.25 |
#12 | 1.75 |
#13 | 1.25 |
#14 | 0.75 |
#15 | 0.375 |
#16 | 0.0 |
All of these (16) items are a measure in fluid ounces of how much hydrated aluminum solution remains in the container AFTER each addition. The last one (#16) is zero because prior to dumping in #16, there is .375 fluid ounce left in the jar and addition #16 is the pouring in of that amount, leaving the container empty. That is shown on the sheet as "All In".
An additional note about these portionwise additions is that if you want to know what the individual additions are, simply subtract what's left over after each one, from the amount above it.
For instance: It starts with a total of 10.75 fluid ounces. After splash #1, there are 10 fl oz remaining, so that first addition is .75 fluid ounce. You can go down through the whole table then, and translate each addition into its exact amount. To check them, make sure they all add back up to 10.75 fl oz
From there you could go on to translate from fl. oz. to milliliters and then use a pipette, burette, or any other convenient way to add those (16) precise amounts at the correct times, at the correct temperatures.
I just have been using the plastic jar, the string, and the measuring cup as the easiest, simplest, most available way to get it done reasonably accurately. (see the waxmaking video) However, with the information given here, many other methods are possible.
WORK WITH "BATCH SUMMARY"
See the batch summary in the "Tech Musings" section of this website for additional reference.
A FEW GENERALITIES
Some overall generalities which apply to making this type of brown wax are:
1) The amounts of hydrated aluminum added throughout the cooking tend to be larger near the beginning, and smaller near the end. The temperature of the melted stearic when each addition is made tends to increase as the batch progresses. The time intervals between additions tend to increase as the batch progresses. However, this is not strictly adhered to and it has been found that certain patterns of added amounts at various times and at various temperatures throughout the cooking will produce quite noticable differences in the final product. I call this the "Time/Temperature/Amount" profile. Many varying profiles have been tested, and this one outlined here on this batch sheet has been found to get the best results.
2) The general trend is to heat the melted stearic to at least 190 C before adding hydrated aluminum which is at room temperature. Upon adding the hydrated aluminum solution, the temperature drops down to around 130 C as the water boils off, and as the sodium and aluminum stearate forms a precipitate. The temperature gradually rises as the water gets boiled off. It continues to rise after the water has fully boiled off. It is during the time after the water is gone, that the precipitate dissolves in to the melted stearic. As more and more precipitate is dissolved in, the saponified stearic starts to turn color. First yellow lemonade color, then later more of a brown. As the stearic saponifies, higher temperatures are seen at the point when all the precipitate is dissolved in. Less heat is required. The reaction starts to become exothermic. So much so, that later on during the high-temperature end of the cooking (220 C, 250 C, 270 C) the batch can run away and go off-scale high temperature if one does not closely monitor the heat and the temperature. The high temperature of 270 C is needed, but also the situation requires care not to allow it to run away much higher than that.
3) The goal of each of these temperature rises is to use just enough heat so that when the temperature reaches the specified number on the schedule, the elapsed time will match that specified on the schedule. The ideal goal is to have the precipitate fully cooked in by the time for the next addition and this is supposed to happen at the temperature shown on the sheet. This of course cannot happen perfectly each and every time. When the batch does start to wander off of schedule slightly, it is better to meet the temperature exactly and allow the time to go "late". If too much heat is used and the temperature arrives too early, one is forced to make a quick decision "on the fly" as to whether to just add the next addition and keep going, or to hold off and let the temperature go high and continue cooking. It has been found that missing several time points, being late by a few minutes, and also being high on the temperatures by 10 to 15 degrees C does produce inferior brown wax.
It works better to be somewhat easy on the heat so that the temperature cannot ever be in error more than maybe 5 degrees C at most if it does go over. Typically, it has been found that times can be allowed to drift "late" by up to a minute or even 2 minutes in certain places in the schedule, and by so doing, the temperature points in the schedule can be met exactly. So, this is a way of maintaining temperature always very close on schedule, by allowing time to drift late. Once time does drift late, then applying a very small amount more heat will bring the batch back on schedule within the next few additions on the sheet. This is well illustrated on the example sheet. Throttling the heat is the method used to "steer" the batch, and to keep it on schedule.
The drifting of temperature is not well shown on the example because the particular wax batch shown just so happened to have all the temperatures work out exactly correct. More often though, at least one or two of the temperatures will drift 5 or even 10 degrees C high, which gets noted on the form as it happens.
In general, as long as everything ends up at the proper time and temperature at the end of the cooking, with the batch right on schedule, it's no big deal if a few times and temperatures are "busted" along the way in a few places. Time can drift late by up to 2 minutes. Times should not drift any more than 2 minutes at most, because if it gets much later than that, it's quite difficult to bring the batch back on schedule without seriously busting the next temperature way high, and that will produce inferior wax.
4) Ideally, the precipiate will be fully cooked in when the time and the temperarure arrive for the next addition. This cannot always be exactly realized. Most of the time it can be. But sometimes some small amount of the solid precipitate will still be floating around when the next time/temperature point on the schedule arrives. This is where experience is the guide. It has been found that if only a few small specks are floating around and are not fully dissolved in, it is ok to add the next hydrated aluminum portion. However, this trend will build up and amplify itself if given the chance. If more and more undissolved precipitate is seen at subsequent additions, this can lead to a poor wax batch. If this starts to happen, more heat helps get that precip dissolved in as it should be. But, in so doing, then one must watch very carefully that the next temperature point does not get exceeded, nor the time allowed to be busted by being way too early.
In an extreme case where lots of extra heat must be used in order to dissolve in some very stubborn precipitate, increase the heat way up and stir. Make sure the temperature never exceeds the next point in the schedule. If the temperature gets up very close to the next listed tempertaure in the schedule and several minutes early, with solid precipitate floating around, then the container can be removed from the heat, stirred, and allowed to cool for half a minute or so. Let the temperature decrease 20 degrees C or so, then start again on the heat, stirring, and continue.
In this way, stubborn precipitate can be fully cooked in on schedule without ever busting the next temperature on the sheet. If it is taken off the heat and allowed to cool too much, then it is also very easy to get way behind schedule and then it is very difficult to catch up again.
5) This method of having a schedule to follow to help each batch be consistent should naturally tend to maintain itself. Since this schedule was arrived at using the same equipment as has always been used here, if a moderate amount of heat is used, it just naturally tends to arrive at the temperatures and times and with precipitate fully cooked in as shown on the sheet. This type of method should have inherent stability, and not be difficult to adhere to. There will, of course, always be some minor corrective action required, and small adjustments made to the amount of heat applied along the way, but these should be relatively minor, and the trend should be one of the whole works more or less following the sheet without extreme effort.
6) By following this general method, it is probably possible for anyone to come up with their own ideal schedule which will be suited for their particular batch size, and their own unique equipment. The examples shown here are unique to this particular batch size, with this particular set of equipment, and these particular brands of ingredients. Changing ANY of the above will change the outcome.
A LINE-BY-LINE LOOK AT THE BATCH SHEET
As the cooking of the wax batch progresses, the sheet gets filled out. Having a straight edge positioned under each line on the sheet as each step is done, and then moving it down to the next entry helps in seeing what has been done, and what is next. I keep an accurate clock close by the wax as it cooks, and another clock at the table where the sheet is, so that the actual times can be filled in on the sheet as the batch progresses.
Looking at the blank form starting from the top, first the batch number and date are entered and just below that are some ambient temperatures and a little weather report. On the right side at the top is a short description of the type of hydrated aluminum being used. Below that on the right side are a few reminders about the setup.
The first time to be filled in is "light stove", and you see that happens at 11 minutes after a given hour. The minutes are part of the blank form, but the hours are left blank to be filled in when it is known when the starting time will be.
Usually I'll be doing the final filtering of the hydrated aluminum (called H/A on the sheet), then measuring it out at 10.75 fl oz into measuring jar "D" and noting how many times this jar has been used. Then I make sure the stove is fueled up and that the 211.8 grams of stearic has been loaded into the cooking pot, and that the weight has been checked. Only after knowing that everything is ready do I go ahead and fill in the hours into the blank schedule.
You will notice that the stove gets lit at 11 minutes after the hour and the batch is started cooking 5 minutes later at 16 minutes past the hour. This is to allow sufficient time for the gasoline powered camp stove to get up to temperature burning clean.
A glance at the filled-out sheet for batch #112 shows that this batch started with lighting the stove at 10:11 am, and the pot was set on the flame at 10:16 am.
All times in brackets such as [start 10:16] or [10:25], designate the target times. Just to the left of the bracketed target times is the total elapsed time. You will notice that the time for the first addition of hydrated aluminum is targeted for [10:25] and that is at the elapsed time of +09, which means 9 minutes after the pot was set on the fire at [10:16]
So, for the first 9 minutes the solid flake stearic gradually melts and once in a clear liquid state, the temperature begins to rise. If all happens on schedule as it did during batch 112, the time at which the melted stearic just reaches 190 C will be at or very close to 9 minutes elapsed time. For batch 112, this happened 10 seconds early at 10:24:50, which you will see filled in under the word "Time". Then, over on the right you will see that the temperature met the expected target of 190 C and that actual working temperature was entered under "Temp. C"
As stated before in this text, batch 112 was very lucky in that all the actual temperatures matched the targets. Usually one or more will miss slightly and those are noted and a curved arrow is drawn pointing from the target to the "busted" actual temperature which is in slight error from the schedule.
So, right at 10:24:50, the first addition (splash #1) is poured in and the pot is again covered. After entering the data on the sheet then splash #2 is measured out, checked, and covered for use 5 minutes later at [10:30] which is at a total elapsed time of +14 minutes.
Again, the temperature is watched closely as is the cooking-in of splash #1. It just so happened that splash #2 worked perfectly, meeting both temperature and time exactly at 10:30:00 at 190 C.
Then measure out splash #3, check it, cover it and watch #2 cooking in.
You will now see the batch start to drift a bit for splash #3. Note that the target time for it is [10:34], but that it did not go in until 10:35:00 at 190 C. So, splash #3 met the temperature exactly at 190 C, but by waiting an extra minute and going late in order to be at that temperature, not some cooler temperature which it was one minute earlier at the target of [10:34].
As this happens, one thinks to one's self: "We met the temp ok, but we are now down one minute!...better get back on schedule"
Now watch what happens on splash #4: It is met at 190 C exactly, but late by 2 minutes and 20 seconds! That is late enough to cause concern because it exceeds 2 minutes. Better add some heat and get this back on schedule!
Splash 5 shows that we are catching up now after having added a bit of extra heat. Again we hit the target temp exactly, and we have gained a bit of time, being only 1 minute, 20 seconds late now. Better.
Splash #6 is better yet because the temp is perfect at 190 C, and we have gained more time, now only being late by 30 seconds. (Almost back on schedule!)
Splash #7 is more like it! Hit the temp at 200 and only 7 seconds late now.
After splash #7, the heat must be reduced in order to hit the next temp of 190 without going over.
Splash #8 hits it just right at 190 C and now 5 seconds early at 10:56:55 for the target of [10:57]. Not bad. 41 minutes in and completely back on schedule with no busted temperatures.
Splash #9 then goes 20 seconds late again, but meets the temp of 205 C just fine.
Splash #10 makes up that late time and hits everything on the mark. 52 minutes in and the batch is on schedule.
Splash #11 hits another bullseye. This marks the transition point between the early low temperature portion of the cooking, moving on into the high temperature range toward the later stage of cooking. It is good to be on schedule as closely as possible at this point due to the fact that from here on out, the precipitate becomes noticeably more difficult to fully dissolve in. Splash 11 foams up very tall in the container. It usually gets up to within an inch or so of the top of the pot as soon as the lid is removed to start stirring it down and cooking it all in. This is why a tall, narrow container is required, because at certain points during the process, the contents goes almost completely to an "all foam" condition and if the container is too small, it will blast the lid off and boil over. Notice also that the temperature for splash 11 is 220 C.
Splash #12 drifts 2 minutes and 45 seconds late again, but with the temp fine at 220 C. (getting ready now for the high temperature end of the cooking)
Splash #13 makes up some time and is only a minute late now with the proper temp of 250 C.
Splash #14 makes up the remaining minute and is on time and at the full 270 C.
Splash #15 is 20 seconds late, but right at 270 C.
Splash #16 happens 10 seconds early at 11:55:50 instead of the target time of [11:56], but at the full 270 C.
Not bad, 1 hour and 40 minutes total elapsed time, and on schedule within 10 seconds. And early to boot! Early is good at this stage.
However, we now see that after splash #16, ideally, all solid precipitate should be cooked in because the sheet says "All Cooked In" at a target time of [12:05] which is 9 minutes after having added splash #16. Did not quite make it this time, as there were still a few chunks floating around needing some more cooking in. So the words "not quite" are written in before "All cooked in" to show what actually took place at target time [12:05]
That means there are now 4 more minutes of emergency "pad time" which can be used to do this extra bit of cooking. I was too busy doing that at the time, but later on wrote in "use pad time until 12:09 to cook it all in". Notice that the target temperature shown for [12:05] is a range from 260 C to 270 C. This provides a reminder that a bit less heat is needed if all precipitate is completely dissolved in by [12:05]. In this case it should be allowed to simmer at 260 C for the 4 minutes before adding the last portion of stearic. If however, there is still solid precip floating around at [12:05] as was the actual case with batch #112 as shown in the example, then the higher temp of 270 C should be used. In short, there is no need for excess temperature here, if the precipitate is all finished dissolving in. Easing up on the heat here helps maintain very quiet wax.
Next entry on the sheet is "+1 OJ stearic" which means to add 70.6 grams of stearic. At this time of [12:09] all of the precipitate MUST be cooked in, and that means that the work during the last 4 minutes has used whatever heat was required to get it done, and with lots of stirring. This was accomplished by going no higher than 270 C, but a few times in the past, 275 C has been used and even once 280 C, but those temperatures are rare and very seldomly required.
The solubility increases as temperature increases, so this can be used to help dissolve in all of the precipitate.
After the 70.6 grams of room temperature solid stearic has been added, that causes the temperature to decrease markedly. This added stearic must be melted in, and the whole heated in order to cook it in for ten minutes. This ten minute cooking-in period starts as the mixture passes through 232 C. The sheet says "Start 10-min stearic cook-in" and then "Passing through 232 C", where you will see the time of 12:12:30 filled in. The target time for that is [12:13].
During this ten minutes, the batch is heated to 250 C and taken off the heat and stirred thoroughly, then taken under a 100 watt lamp to very closely examine for any tiny small precipitate bits floating around. Then it is put back on the heat and again taken up to 250 C for a 2nd time, and again taken under the inspection lamp after a good stirring. At this point it should look just like a hot cup of the blackest coffee you have ever seen. There should be absolutely no particles floating in it. It is a very dark, very thin watery liquid at this point. The reason for the two temperature excursions to 250 C with lots of stirring is to make absolutely sure that all of the precipitate is fully dissolved in. The two extra trips up to 250 C provide two last and final periods of increased solubility to make sure that gets done.
It is then allowed to cool down to 232 C, which is shown to have happened at 12:18:30. Once back down to 232, it stays there at that temperature until the next step at [12:24] which is to add 17.3% ceresin. That happens to be 59.1 grams of ceresin. From this point onward, the wax batch never gets heated beyond 232 C. The high temperature period has been completed.
The adding of the ceresin is targeted anywhere within the minute of [12:24] to [12:25]. This target covers a whole minute because getting the ceresin in there without making a mess and causing hot wax to splash out takes a bit of care and time.
There are now 7 minutes in which to melt the ceresin in and get the batch back up to 232 C before the start of the ten minute cooking-in time for the ceresin which begins at [12:32]
The ceresin cooks in for those 10 minutes, being gently stirred a few times. The batch is finished by [12:42]
Total elapsed time is 2 hours, 26 minutes.
A 1.25 teaspoon sample is poured off for later testing.
The time is noted. 5.5 hours later after the batch has cooled and hardened, it will be ready for the first aging step.
That finishes brown wax batch #112, made 6/16/15.
THE AGING OR "CONDITIONING" OF THE WAX
Refer now to the back of the batch #112 sheet.
This shows the (2) aging steps which need to be done to this wax in order to make it ready for casting. Also shown after the aging steps are the castings themselves, which made blanks M337 and M338.
It has been found that although this wax can be cast into blanks immediately after having been made, that the odds of making decent castings which yield up good quiet blanks are much better after this wax has been melted and hardened a few times after having been made.
Statistics show that casting the wax right away tends to produce noisy blanks more often than casting it after it has been conditioned. The conditioning goes a long way toward increasing the odds of success.
So, first it is re-melted and raised to 232 C and held there and stirred gently a few times for 20 minutes.
Then it again is allowed to cool and harden for 5.5 hours minimum time.
Then after hardening, it is again heated to 232 C, stirred and held there for 5 minutes and allowed to cool for 5.5 hours minimum.
These (2) extra steps are worth doing because by doing them, one avoids casting some noisy blanks.
For added purification, each time this wax is allowed to cool and harden, the top surface of it is scraped clean and the resulting fine dust is vacuumed out. This dust contains a few microscopic aluminum particles.
We loose a bit to vapor during these steps. This time we lost 8.5 washers to vapor. (.2674 x 8.5 = 2.27 grams lost during aging)
Following along on the sheet, it can be seen that there ended up to be 1228.5 washer weight available to cast.
Then casting #337 used 459 of that, for 769.5 left after casting.
Then casting #338 used 450.5 of that, for 319 left after casting.
Then the remaining leftover wax chunk was taken out of the pot and weighed separately and that weight was 318, for a weighing error of 1 washer, which is .2674 grams. Not bad considering that we started with 328.5 grams of wax to cast.
Those left-over wax chunks from various batches are saved and used along with the cutoff ends of castings to make the scrap run blanks described in full detail in the "Tech Musings" section of this website.