Making records circa 1979 Pitman NJ

[ljc300]

Sorry for the long post, but here goes:

This post was inspired by the videos on Youtube of how records are now made. They are pretty close to what I experienced back in the 70s where I worked at the CBS plant in Pitman NJ in the electroplating and "matrix" dept. The plant produced about 250,000 records per day. I since got a chemistry degree and I work in that industry now, but my interest in vinyl records persists. The videos may be the way records are made today, but older record collections reflect the manufacturing practices that I participated in up to 1980. I've forgotten some details, but hopefully what I remembered will be of some interest. I also welcome any questions it may conjure up.

The relevance of this post may be for vinyl-philes to gain insight into where the noise comes from in records, and why they are often great, usually good, and other times neither. Or from my point of view, how amazing the sound they produce is considering the technology that went into them, at least compared to today's standards. Lastly, I'm glad that people are still interested and spend thousands of dollars on playback equipment to capture their information.

Here are the youtube videos I found by simply searching "how vinyl records were made" in google:

http://www.youtube.com/watch?v=IReDh9ec_rk&feature=related
http://www.youtube.com/watch?v=xUGRRUecBik&feature=related

In this discussion, the sequence of steps from the lacquer to finished record is as follows:

Lacquer->Master->Mother->Stamper->LP Record

These numbers are a bit sketchy, but as I recall, a few masters could be made from a lacquer, perhaps a dozen or more Mothers from each Master, and many dozen Stampers from each Mother, then perhaps hundreds to a thousand records from each stamper. So you can see that one lacquer (which had to be perfect) resulted in many thousands of LPs. Hence, only a chosen few were trained and trusted to handle lacquers...

After the nickel "Master" was made, you had a mirror image of the record, i.e. "humps" and not "grooves". It cannot be played on a turntable. In the video, this part was called a "Stamper". Though this may be the practice today, we never took these directly to the presses; it would have been WAY too slow and expensive. You only got a limited number of records from each stamper (~1000 on a good day), so it was cheaper to use the Master as a template to mass produce MANY Stampers by adding two additional steps.

We received the precut lacquers from the NY studio. Our CBS plants in Pitman NJ and in Terre Haut IN made records of just about every label except for RCA and Capitol, presumably they had their own plants. Computers weren't around back then on the cutter, it was completely analogue. Because of the expense and care in making them, these laquers were the "gold nuggets" or "unfinished diamonds" of the factory, and were treated as such. We'd first spray them with a solution of silver nitrate/surfactant, then with a sucrose solution. (We didn't use the tin method described in the video.) The sugar reduced the silver from the dissolved nitrate salt to silver metal which adhered to the lacquer, much like making a mirror. The surfactant got the silver solution deep into the grooves, otherwise it would just bead up. The lacquer then had a micro-thin conducting surface, required for further electroplating of nickel. The nickel was necessary because it is more durable for the subsequent machine operations. This is shown in video 1 where the silver coated lacquers are immersed into the green NiCl "Watts" bath which was virtually identical to the equipment I used.

Nickel was "rented" from International Nickel Corp, that is, all parts were returned to them, and converted back into 1-2" nickel balls for us to use. These are seen suspended in cloth bags dipping in the plating bath. As the electroplating proceeded, these balls dissolved while nickel plated out on the part. The part was the cathodes and the nickel balls the anodes. This was all done at low voltage but high amperage. The operators had to keep a close eye on the volt-ohm meter, if the voltage went too high it meant that the resistance was too high. The circuit was a crude galvanostat that kept the amperage at a high value to hasten the electroplating, but the downside was that if the contact was poor or deteriorated, the device would compensate by increasing the voltage. The resulting ohmic heat could and did destroy the submerged part. The parts made electrical contact at the very center, and the copper washers used to make that connection had to be clean and make VERY good contact. The bath was "supersaturated" in the nickel chloride, that is, heated up to permit very high concentrations of electrolyte. If it cooled it, for example, on your skin, it immediately crystallized out. The batch was circulated through a large system containing filters and a carbon bed to remove organic impurities. Many defects found on parts had been attributed to the plating bath. The staff chemist, affectionately known as "Mr Wizard" (he had the highest technical degree in the plant, a BS in chemistry), forgot more about electroplating than most people ever knew, at least from the practical side. One even that occurred during my short tenure was defects known as "pools". These were small, oval circles, perhaps a few mm in size that began showing up around the outside edges. The analytical techniques of the day were not able to find the bath impurities, if indeed they were causing them. Finally, the chemist and one of the more seasoned mechanics traced the problem to mercury switches that were used on the automated plating machine. The plating bath was very corrosive, and the switches were replaced often. Ultra trace amounts of mercury escaped into the bath and that was found to be the root cause. Unfortunately I don't recall the remedy, but removing trace amounts of mercury from a supersaturated nickel bath would have been a difficult and horrendously expensive problem to achieve! My guess is that the entire batch had to be replaced.

CBS had a special designation for their higher quality classical recordings, called "Columbia Masterworks" series. In the electroplating dept., these were distinguished from the pop recordings by using a much slower electroplating step during mastering. Apparently, slowly building up the nickel layer produced superior product, otherwise they NEVER would have wasted their time and money. They were always looking for ways to cut costs. Subsequent to that, I don't recall any difference other than the way they were packaged (better sleeves) and that they contained more vinyl to resist warping.

One very important step was omitted in the video that I'm sure is done today: All to-be-plated parts (including the de-silvered master from the original lacquer, more on that below) had to be "passivated". In that very critical step, the part was dipped in a caustic solution and anodized (set to a positive voltage) for a few minutes at a low voltage and current. This put an atomically-thin metal oxide layer on the surface. That way, after the nickel layer was added the two could be separated. Imagine a few atomic layers of metal oxide sandwiched between the two metal parts. The passivation step was done on all the parts between their respective plating operations. The separation step was all done by hand, like peeling the backing off of a sticker, even though the edges of the parts were razor sharp. You had to be careful to not add a wrinkle, dent, or scratch, else it became scrap. With practice and diligence, even minor cuts became very rare (more on this below...).

Masters had to be "de-silvered", not so much to recover the silver that adhered to it, but to get a pure nickel surface. That was done by a 'reverse plating' type step, where the part was anodized and the silver stripped from the surface. This took about 30 seconds to a minute, but not too long, or you start to strip off nickel and hence, music! I don't recall the consequences of skipping the desilvering step, but it was very easy to spot a Master that had the silver still on it. My guess is that if it were passivated, it would be desilvered too. Both of these steps were done in separate baths to keep a smaller volume of silver solution available for recovery.

When certain operators forgot to do the passivation before plating, the subsequent parts became permanently fused and were tossed in the recycle bin. This rarely happened, but when masters-mothers were involved, it was the worst case because you could only make a few masters from a lacquer. Not good!

It is intuitive that the first Mothers produced from a Master (over a dozen could be made before the Master inherited defects) might make the best product. If you look very closely at the center of a vinyl record, you may see a very faint capital letter stamped on the opposite side of the serial number. It was stamped on the Mother by the plating operator. This was how we kept track of how many Mothers were made from a given Master, or put another way, the number of times a Master was used. The letter "A" means that the Stamper was made by the first Mother made from a Master. The letter "D" from the fourth Mother, and so on. It was rare to get more than 20 Mothers out of a Master, so the alphabet had plenty of characters for this system to work. Since records were the next audio QC check, the pressroom testers had to tell the matrix dept. when the Stampers were off quality, at which point the Mothers were rechecked. There may have been a default number of Stampers that were produced from each Mother, because at that point you were reaching a point of diminishing returns.

Though Masters cannot be played back (have humps vs. grooves, Mothers can be. Mothers were polished with rouge, similar to what jewelers use (probably the same thing). It looked exactly like automotive rubbing compound. The idea there was to remove microscopic burrs that were a result of the electroplating, or other mechanical defects, so that they could be audio tested. In fact, all Mothers were tested by human listeners, using turntables and scads of styli that quickly wore out rubbing on nickel. As a footnote, the testers were all gals, and at the time the feeling was that women possessed better high-frequency perception than men. New employees were screened upon hiring using the best tests then available, and those with the best hearing were sent to the listening booths. Occasionally, a tester would have a Mother polished additional times until the sound was right or all the pops and clicks were removed. When a defect was found in the form of a "pop" or "click", it was often repaired under a microscope by a skilled engraver. It would then go immediately to the polisher/tester phase and the tester would listen to the spot that was repaired. This was a crude but cost effective way of salvaging what had already been invested in that part.

I still have a few samples of these. When you play a lacquer, the dynamics are considerably different than a finished record, but INCREDIBLY rich sounding. That is, the loud parts are REALLY LOUD and the soft parts about the same. (I have a lacquer of a Cesar Franck piano/cello sonata, and it's simply an amazing listen...) I'm guessing that lacquers were cut to take into account the "information loss", or at least the "shift” that occurs during the chain of electroplating operations, or to the mechanical differences that are inherited by each part in the chain. The sound that came out of Mothers was VERY hard to listen to: brash and noisy, presumably because their surface is "crystalline metal” and not smooth like vinyl.

The Stampers couldn't be tested directly (humps again...). The hole-centering machine on the video is exactly the same one I used, very crude. You literally spun the part while looking at a small segment of the "true center circle" through the magnifier until it stopped moving side-to-side. There was a knack to "whacking" the Stamper with as few as hits as possible until it was centered. We actually made a game of it. An adjustable felt gripper initially secured the Stamper, which was gradually tightened as you got closer to center, then once perfectly aligned you clamped it down tight and punched a true-center hole. A skilled operator could do this in two to three hits. During backlogs these guys were called upon to get us "caught up". The interest here to audiophiles is the occasional "off center" record. A really bad one makes you nauseous while listening to it, and eventually wears out quicker, given the period shift in skating that occurs while it's playing. But as crude as the centering step was, off-center records were statistically very rare, as the pressroom testers caught them from the get-go.

Once the center hole was cut, the Stampers went on to get various machine operations to make them fit the presses (beveled edges, crimped center hole, etc.). Needless to say, all of this work was inherently dangerous because up until they were trimmed, the edges were razor sharp. I witnessed a worker get the top of this thumb cut off when he was polishing a Mother with rouge on the spinner. Imagine a 200 gram 12" diameter razor blade spinning a few hundred RPMs, and polishing it with a cloth from center to edge with your bare hands ! Surprisingly everyone respected the situation and cuts were relatively rare. It would be interesting to see how these operations are dealt with under today's OSHA rules.

To protect parts from air oxidation between stages, they were placed in wooden stackable trays after being coated with oil mixed with soap. Before mounting on the plating stages, this coating had to be removed with water, then steam, and the part polished. Otherwise, the pure nickel surfaces (and it was PURE!) would get defects on exposure to the air. The environment was anything but a "clean room" affair used in the electronics industry. Masters and Mothers were stored long-term by coating them with wax. When an archived part was needed, the wax had to be carefully removed with heat and solvent. Hence the importance and value of original releases of recordings, rather than re-releases. It was luck of the draw to get one that was made from a new lacquer or from an old Mother that was taken out of storage.

Stampers made it to the presses before oxidation became a problem so weren't coated. They were mounted on the presses and the first few records and several sampled at intervals were tested just like Mothers. This time it was the final product being listened to, and it was a very closely watched activity. Down time on presses = $$$! Occasionally a press would lose temperature, seize up, and be a mess to fix. The press mechanics were some of the most highly skilled workers in the plant, since everything could shut down at both ends of the process with no backup work to fill the time.

The vinyl used for making LPs (polystyrene was used in the injection mold machines for 45s) was purchased as a fine white powder, and the incoming trailers were tested for various properties such as volatiles, moisture content, and viscosity of a PVC-solvent solution. Though the testers weren't aware, they were actually measuring the MW distribution of the PVC which was ultimately responsible for its melting point, strength, and flow characteristics under pressure. The white powder was blended with carbon but I don't recall any other additives. So, all the talk about plasticizers and special mold release additives, etc. may be pure speculation, but I only worked for one company, there may be many ways to tackle this. However a simple analysis of a chunk of record would put that argument to rest (I haven't done that yet, and don't have any I want to sacrifice!)

[Aussie0zborn]

Welcome to the forum, LJC and thanks for your your first post. Not much has changed in the process.

Are you sure the silver plated lacquer disc was passivated prior to plating?

[ljc300]

No it was not, the silver was plated directly with Ni but then the Ag was stripped off after peeling it off the lacquer. Then it was passivated before adding more Ni.. The passivation was always done before plating Ni on Ni so the parts could be separated. Sorry for the confusion.

[montalbano]

Hey LJC

Like Mike said, not much has changed in the process.

About the hurting you mentioned during the polishing process, nowadays you have 2 chances:

- 1. use a magnetic polishing machine, we have here a Capitol Records automatic polishing machine which does the whole process. FYI it works A OK, but we still prefer to do the job manually.

- 2. use a manual backsander, cut one thick piece of rubber and put the emery below it while you are backsanding. The rubber piece will prevent any hurt. You will first use a rough emery and later on a finer one. Once you will feel that the surface is smooth and clean, then the stamper will be ready. For best results and no damages, you mount the stamper on the backsander, let it spin, and operate +45° from the top. From center to edge, continuously.

The final check is visual and you can still correct any slight imperfection by hand with some emery, just before mounting and do the final cleanup with acetone, again visual inspection and eventual air blowing.

BTW you mentioned the backsanding of a mother, but there is no need to do such a thing because what you effectively need to backsand are negatives (eventually) and stampers.

As a final suggestion I can add the following:

- never let a stamper form in 2 separate times, e.g. overnight and then the day after. It will delaminate. Stampers have to be formed continuously. Bath's temps 45°C or 60°C will be OK regardless, you just have to supply the proper Volts which will result in the corresponding amperage according to bath's temp and capacity. Bath's pH has to be around 3.8 for best results.
- it is a good idea to backsand the stampers or the negatives before their separation; this way the backsander's surface will hit against the back of the stamper or the negative. And you will have another advantage: that - by using a thin holding pin in the center - you will be able to backsand an extra surface very close to the center. Once you will center the stamper, a 1" diameter will be cut out, so if you had backsanded to the very center, then you will be able to preform and supply a 100% clean stamper to the pressing dept.

This is my experience and what we do here, but of course someone else might have come to most advanced techniques.

[ljc300]

How could I have forgotten the most exciting part of stamper making! We used to take the stamper after it came out of the bath, mount it on a hand-held "flywheel" for lack of a better term, and while pinching the stamper with a big leather "mitt", push it against a big belt sander (about #50 grit). With a lot of finesse you got the burrs off the back of the stamper without burning it up, grinding it too much, or cutting your hand off! I've never seen anyone get hurt doing this job, people respected it as probably the most dangerous task in the shop.

I may have confused folks somewhere along the line in my post, I didn't mean to imply that we sanded the back of all the parts, all I remember was polishing the actual playing surface of the mothers with rouge, and sanding the back of the stampers. Otherwise, the little bumps would actually come through on the presses onto the vinyl.

thanks for your reply.

[Aussie0zborn]

Hey LJC, can you tell us more about polishing mothers with jeweller's rouge. This has been mentioned here but I dont think anyone actually polishes and de-clicks mothers by hand anymore - its probably become a lost art.

Some detailed info for those who might want to de-horn and de-click mothers by hand might find it helpful. Thanks in advance.

[ljc300]

The mother is mounted on a wheel that spins at several hundred RPMs, vertically, there's a vented shroud on it to catch debris and liquid that flys off during the operation, sort of like a bicycle wheel fender. There are water and steam nozzles available at the workstation. All the operations happen with the mother spinning. You had to be EXTREMELY careful to mount if so it wouldn't fly off and when your fingers go near the edge, well, it's like a meat slicer without the guard on it! The stations were spaced in a safe manner and people were really careful and respected each other's safety.

1. Spray the mother to remove debris and the oil emulsion, then hit with steam to dry it off.
2. Moisten a cotton pad with water and a small amount of rouge (looks just like auto rubbing compound, not sure of the "grit")
3. With moderate pressure, press on the mother from the center towards the edge in a sweeping motion so as to evenly and smoothly polishy the entire grooved surface.
4. Alternately spray with water and steam to remove the rouge.
5. reapply the emulsion protectant unless the tester needs it right away.


The Emulsion I refer to is a mixture of some kind of oil and soap so that it is easily removed with water. It's just to keep the Nickel from contacting air.

I've not done the actual mechanical repairs under the microscope, but the gals that did that could save a lot of parts. It was usually an iterative process:

polish-->listen--polish-->depop--polish-->listen etc. We'd usually give up after a couple of rounds of that without success.

[dubcutter89]

This has been mentioned here but I dont think anyone actually polishes and de-clicks mothers by hand anymore - its probably become a lost art.

Hmm, I think Pallas in Germany does it... but I'm not sure...

Here's a little report about them (in Deutsch)

http://www.analog-forum.de/wbboard/index.php?page=Thread&postID=509444#post509444

Lukas

[mossboss]

Hey all
The post from Pallas it's interesting for sure
My deutcsh sucks but I can make out that this may be a Stanton System with the addition of a projecting microscope as shown on the screen it would work well
This guy is listening and marking for manual de clicking rather than what is described with rouge
This is a very selective procedure where the spot is marked where there is a pop or a click than it is removed with a sharp instrumnt like a sewing needle whil looking at it under the microscope very neat and tidy and no rough rouging either
Common practice in the days when the runs where reasonable but applied by very few today and to be honest with the average run of say 300-500 units not quite justified cost ways any way
It may take a whole day som times to do a decent job on a set of mothers It can never be faster than the length of the tracks plus another hour which is a two hour ex raise any way so cost does add up
By the way
If people are prepared to pay for a three step process plus the time that it takes to de lick and de pop a mother we do do it and it is not a lost art at all
What the other poster describes is a very rough way of de whatever the stamper or mother which we do under duress or reluctance
We will only do it after hearing the pressing I can assure everyone there is a dullness in the sound introduced into the records after that procedure so we avoid it prefering to re cut rather than compromise the sound
Of course some people today don't care they just want something to sell so it becomes a case of if that's what they want well
Procedures from emi that I have in my possession describe a similar but quite precise procedure for treating mothers prior to manual de clicking but with very specific " rouge" obtained by some supplier