Recording for Vinyl

Vinyl record sales have been increasing every year for a while now. Today, vinyl outsells CDs. Fans like vinyl for its unique sound, and the large format of the jacket.

Record pressing plants are busy! Most are backlogged for months.

The process of converting your digital mix to a vinyl record is complicated. The entire system of phonograph records is plagued by many deficiencies, which makes cutting the master lacquer disc a challenging art.

How we record and mix has impact on how good the vinyl record will sound. In this episode, I explain some of the fundamentals of the master disc-cutting process and provide some insight on how that process will change the sound of your recording.

I explain some factors in recording and mixing that will result in better-sounding vinyl records. These procedures are not complicated, but understanding these techniques will help you and your clients get the most out of the vinyl medium.

email: dwfearn@dwfearn.com
www.youtube.com/c/DWFearn
https://dwfearn.com/

Transcript

101                        Recording for Vinyl                                              January 21, 2025

I’m Doug Fearn and this is My Take on Music Recording

Vinyl records are popular. Actually, the upward sales trend has been going on for a while. Listeners like vinyl. Pressing plants are busy, often booked out for months.

Vinyl has a unique sound, which many listeners appreciate. The album jacket is a square foot, plenty of room for large photos and lots of liner notes.

In this episode, I talk about the challenges of making great-sounding vinyl records, and what you can do in the recording process to realize the full potential of vinyl. I am mostly talking about 33-1/3RPM 12-inch, long playing albums, which I refer to as LPs. But most of this applies to 45RPM 7-inch singles as well.

First, a few descriptions to make sure we are all talking about the same things.

1.       The master disc has an aluminum base that is coated with a precision layer of lacquer. That is why they are called “lacquers.” Some people call them “acetates” but that refers to an entirely different (and long obsolete) disc technology.

2.       The master disc is cut on a special turntable called a disc lathe. The disc lathe supports a cutting head, which is driven across the record by a mechanism that determines the spacing of the record grooves.

3.       The lathe itself is large and heavy, to support the precision machinery and isolate the lathe from structure-borne vibration.

4.       A considerable assemblage of electronics is required to make the lathe work. This controls the movement of the cutting head and provides the high-power audio needed to drive it. The audio source most commonly is a computer with a digital audio workstation that plays back your digital master.

If a recording project you are working on is going to be released as a vinyl record, do you have to do anything differently from how you normally record for digital distribution?

The answer is that you might get away with doing what you have always done. Lacquer mastering engineers have learned how to deal with the wide variety of recordings, and they have correction techniques to make almost anything work. They are good at it, and usually they can salvage even the most challenging recording and make it into an acceptable record.

But if you want to get the most out of the vinyl medium, you can help the process by making some changes in your recording process so that it translates onto vinyl in the most effective way.

Why do we have to do this? It all has to do with the incredible kluge that a disc recording is.

Phonograph record technology is a mechanical system. The groove on a disc has the waveform of the music cut into it, which you can see even with a magnifying glass. A low-power microscope will reveal even more detail.

The cutting stylus vibrates with the music, carving the waveforms into the lacquer. That master disc goes through additional mechanical processing to make a steel “stamper,” which Is a negative image of the final disc. The stamper presses the vinyl material into a record. On playback, a stylus follows the groove and translates the mechanical waveform to an electrical signal, which is amplified and ultimately heard by the listener.

The vinyl disc medium is one of the few things in the recording world that is intuitively easy to understand, because you can actually see how it works.

This discussion is, by necessity, going to get a little technical. I will try to keep it as simple as I can and give you only the background you need to take full advantage of vinyl. I disregard many details. Those details are necessary to fully understand the process, but not necessary to make an effective recording for vinyl.

What am I talking about? Well, I find it totally amazing that a vinyl record even remotely sounds like the music fed into it. That’s because of the torture that the system imposes on the audio to make it work.

We will start with the RIAA equalization curve. This is a processing technique that goes back to the earliest days of electrically-cut master discs, in the 1920s. It was standardized in the early 1950s. That doesn’t mean it reached perfection, but it is what we have to deal with.

Here’s the problem: the playback stylus following the groove cut into a record has to travel at high velocity across a surface that is not perfect. The vinyl material, no matter how finely made it is, still has texture to it. Under a microscope, it is not a perfectly smooth surface. Magnify it enough and it looks like a boulder-strewn landscape.

All those tiny imperfections are translated by the playback stylus into a steady stream of noise. The noise is loud, and it never stops. Without RIAA equalization, the music would sound like it was recorded with a pink noise generator running at the same time. Not acceptable.

Since the most annoying components of the surface noise are in the higher frequencies, we could reduce the noise by rolling off the top end during playback. But that would ruin the music. It would be dull and lifeless, as if you were hearing it through a wall.

But if we boost the high frequencies, a lot, when we cut the disc, and then apply a corresponding roll-off during playback, the frequency response of the recording is returned to “flat” while the noise is significantly reduced.

This concept is called “pre-emphasis.” The corresponding eq on playback is called “de-emphasis.” Not exactly the best descriptors, but those are the engineering terms used.

Vinyl records are not the only audio medium that is noisy. Tape recorders have the same problem. Tape hiss is terrible, but becomes more acceptable if pre-emphasis is used on recording, and de-emphasis is applied on playback. The noise is reduced by roughly an amount equivalent to the amount of boost. If you boost the highs by 20dB, the noise will be reduced by a comparable amount. It’s not quite that simple, but this explanation will do for our simplified understanding of the process.

Analog FM radio broadcasting uses the same technique, for the same reason. In fact, the RIAA disc equalization, the pre-emphasis used for tape recording, and the one for analog FM all use very similar curves, with about a 20dB boost at 20kHz.

This concept was implemented early in the disc era. But it was not standardized. Each record label had their own equalization curve that they thought was best. If you come across some ancient hi-fi equipment, you might find selectable equalization curves, often labeled by the record company’s name.

Eventually this was standardized and everyone was supposed to cut their records using the same eq, and all the phono preamps were supposed to be designed to exactly match that same specification, in reverse.

Along the way, another equalization curve was implemented that did a similar thing for the low frequencies, except the lows were attenuated during cutting and boosted during playback.

How much boost and cut? The RIAA curve uses 1kHz as the reference frequency. We will call that “0” db. The curve starts rising above around 2kHz and continues to rise until it is +20dB at 20kHz.

The low end is similar. The response is fairly flat around 1kHz and then starts to go down until it is -20dB at 20Hz.

That’s a lot of eq – a 40dB deviation from flat response from 20Hz to 20kHz.

So why is that important for us recording the music? Well, we have to be careful about the high-frequency content of our mix because the highs are going to be boosted – a lot. More than any eq we might ever think of using. If the recording has a lot of highs boosted during recording or mixing, the level could be so high that it will overload the disc-cutting system and that will cause nasty distortion. Additional distortion will be added during the playback process on the listener’s turntable.

To protect the cutting system and reduce the distortion, disc-cutting systems may use a frequency-selective limiter that has increasing sensitivity as the frequencies get higher.

As you might imagine, this can make a mess out of the frequency response for the listener. I have always been amazed that mastering engineers could do this in a way that results in acceptable frequency response, and minimal distortion, in the final pressed record.

It would be better if the mastering engineer did not have to do that. That extreme “air” eq you might love in the studio may not translate as you expect onto vinyl. Better to only boost the highs as little as possible to achieve the sound you want. The result will be better for the vinyl fan.

That’s just one equalization challenge the disc system imposes on us. Another has to do with the nature of the cutting head. This is not something you need to compensate for, but it is a factor in how a record sounds. I explain it here simply to give you some insight into the challenges of the disc cutting system.

In order to cut the groove in the master disc, audio is fed into the cutting head, which you could think of as a strange loudspeaker. Instead of a cone to couple the vibration of the music into the air for us to listen to, the cutting head has to vibrate a sapphire or diamond cutting stylus that engraves the vibrations into the soft lacquer of the master disc.

Imagine that you took a loudspeaker and removed the paper cone. The rest of the mechanism remains, but it is just vibrating the coil of wire that produces the mechanical motion. What would that sound like?

It would not be very loud, since the system has no way to efficiently couple the vibration to the air. And lacking the restraint that the speaker cone provides, the sound would have a huge spike in the frequency response corresponding to the acoustic resonant frequency of coil of wire.

The same thing happens in the cutting head.

In most cutting heads, that resonance is around 1kHz. It is a very high-Q resonance, meaning it is a sharp peak in the response around 1kHz. How high? Typically around 20dB. Imagine how that would sound!

To eliminate that huge peak, a notch equalizer is included in the disc mastering electronics. It must be tuned precisely to the cutting head resonance. You can sometimes hear this extreme boost and cut on a vinyl disc if you listen for it. In the best mastering systems, the anomaly is evened out beautifully and essentially disappears. But there may be some odd phase shift around the resonant frequency that you might be able to hear.

Every cutting head has its own resonant frequency, so the system must be tuned to match the cutting head in use.

The electromechanical system of cutting head, cutting stylus, and lacquer medium has many non-linearities. That simply means that what you get out is not the same as what you put in. To address that, a cutting system uses negative feedback to improve the performance.

What’s negative feedback? Well, the feedback we hear from a live sound system is positive feedback. Without intervention, the squeal of feedback would get louder and louder until something failed from overload.

Negative feedback, on the other hand, works more like correction to any imperfections in the amplifying circuit. All audio amplifiers in our equipment uses some degree of negative feedback. That applies to both tube and solid-state amplifiers. And by “amplifiers” I mean any electronic stage that processes audio. Our equipment is loaded with these amplifier stages. And they all utilize negative feedback to make them sound better.

The negative feedback works like magic, making the frequency response more even – what we call “flatter.” It also significantly reduces many types of distortion.

The downside of negative feedback is that it reduces the gain of the amplifier by an amount equal to the amount of feedback. Also, excessive negative feedback, or a poorly designed feedback system, can alter the sound in subtle ways. So it is always best to use as little negative feedback as possible. It is a tradeoff between the improved performance, the loss of gain, and the creation of artifacts.

In a disc-cutting system, the cutting head has many problems that are helped by negative feedback.

In a loudspeaker, an audio signal goes into a coil of wire in a strong magnetic field and causes motion in the coil that is coupled to the air with a paper of composite cone. It is actually a dynamic microphone in reverse.

The cutting head works the same way, but with many odd resonances and imperfect frequency response. To compensate, another coil of wire exists inside the cutting head purely for the purpose of picking up what the cutting head is doing and sending that as a negative feedback signal back to the amplifier. That corrects many of the problems.

The cost is a reduction in the amplifier output level. In a typical disc cutting system, the negative feedback might be anywhere from 10 to 30dB. As in an amplifier circuit, the best sound is achieved at a compromised point where the errors are mostly corrected but none of the bad things become too audible.

That also means that the power amplifier driving the cutting head has to have a remarkable amount of power output. Most systems use amplifiers from 200 to 800 watts.

What I have talked about so far are just the things necessary to make a disc sound close to the original music. You can see how complex the system is and how many adjustments are available to calibrate it. And over time, those adjustments might have to change, to compensate for mechanical wear within the cutting head, or aging of electronic components.

The lacquer discs are another variable. They gradually dry out and the surface becomes harder over time. The first disc in a shipment of lacquer discs will have different characteristics from the last one. Eventually, the disc is no longer capable of being cut with high quality. Compensations have to be made depending on the characteristics of the disc.

By the way, there is only one continuous groove on a record, although people often talk about “grooves,” meaning the various individual “lines” of audio running parallel to each other across the disc.

The cutting stylus is a precision piece of sapphire or diamond that is ground into a very precise shape. The finer the polish on the surfaces of the stylus, the quieter the groove it cuts will be. To further reduce the noise, the stylus is heated by an electrical current running through a coil of wire around the stylus. It is sort of like the heating element in a toaster, only much smaller.

The heated stylus cuts through the lacquer surface of the master disc more smoothly, reducing the noise level. The amount of heat is critical. Too little and the noise will not be reduced by much. Too much and the heat can actually ignite the lacquer! The “chip” that comes out of the groove as it is being cut is a very fine ribbon. Lacquer is an extremely flammable material, and if the chip starts to accumulate around the stylus, it will catch on fire and burn very rapidly with a lot of heat. That will ruin the disc, of course, but it could also damage the stylus and the even the cutting head. A vacuum system sucks up the chip as it comes off the stylus and carries it away safely. If that vacuum system fails, the mastering engineer has just a few seconds to correct it before the fire starts.

Another problem is that the quality of the audio depends on the speed that the stylus is moving through the groove. That is the same with tape recording. The higher the tape speed, the better the audio quality.

But on a disc, the diameter of the groove starts out much larger than it becomes at the end of the recording, near the label in the center of the pressing. The speed of recording, in effect, is constantly slowing down.

That means that the audio at the beginning of the disc is higher quality than it is at the end of a side. How much different? Well, it’s difficult to quantify in simple terms, but the main effect is a loss of high frequencies and an increase in distortion.

Most disc lathe systems compensate for that by automatically boosting the highs as the stylus gets closer to the end of the side. That helps maintain the frequency response, but that is more high-boost imposed on a system that already boosts the highs by a huge amount.

The audio quality is reduced at the end of the side, and that has implications for the producer or recording engineer when sequencing an album. More on that later.

There is a limit to how long a side on a 12-inch long playing record can be. That is generally about 22 minutes, but like everything else with vinyl, it’s not quite that simple.

If you want to put more time on a side, it requires that the grooves be spaced closer together. And when you do that, you have to reduce the level so that the waveform does not become excessively wide and impinge on the groove before or after. In other words, the more music you put on a side, the lower the level will be. And lower level means higher noise. Another tradeoff.

The “groove pitch” refers to the spacing of the grooves, and could range from 80 to 200 lines per inch. The higher the pitch of the grooves, the closer they become, and that means lower level.

If the level gets too high for the groove pitch, the playback stylus is likely to jump from one groove to the next. The record “skips.”

That can be mitigated to some degree by using “variable pitch.” This is a system that automatically adjusts the groove spacing depending on the level of the music. It’s done by “looking ahead” at the audio and using that information to adjust the pitch. In the days of tape, that was done with a separate playback head, called a “preview head,” spaced the equivalent of one disc revolution earlier than the actual audio going to the cutting head. The same is done today, but it’s done in the digital realm.

Using variable pitch, you can get more music on a side, and the level can be higher because the system spaces the grooves as necessary to prevent them from overlapping and causing skipping.

But much depends on the dynamic content of the music. It works great for many classical music pieces because they often have a large dynamic range with many portions fairly quiet. However, with some types of pop music, the level is always high and there aren’t any quieter sections that allow the grooves to get closer. There is a limit to the ability of variable pitch to increase the amount of music on a side.

Ideally, a side on an LP should be less than 20 minutes. That allows lots of level. The record will be louder and the noise level lower.

Another problem with groove spacing is that a loud section adjacent to a very quiet section may become audible as a pre- or post-echo. The level is way down, but it still can be loud enough to be heard even at moderate listening levels. To prevent these echoes, the mastering engineer sometimes has to go to a higher groove spacing, which reduces the length on a side.

And yet another problem can result from out-of-phase audio in the stereo master. Because of the way stereo is cut into the groove, there is lateral, or side-to-side, motion of the stylus, but also a vertical, or up-and-down motion. That works fine until there is an out-of-phase component to the audio. That becomes mostly vertical motion of the stylus.

The cutting stylus has no problem with that. But the playback system usually has an upper limit on the amount of vertical motion that it can tolerate before the “needle” is thrown completely off the record. It could come back down either earlier or later in the music. Better turntables, tonearms, and playback cartridges are much more tolerant of this, but mastering engineers have to be sure the record will track properly on even the poorest of turntables.

There were always cheap turntables, and although today even inexpensive turntables can be pretty good, you never know what kind of abuse the listener has imposed on his equipment. Even something as simple as the turntable not sitting flat and level will cause problems. We cannot compensate for every possible crummy playback system, but we have to make sure that most people will be able to play the record successfully.

And that means paying attention to the phase coherence of our recordings. Almost all DAWs have a way to monitor stereo phase, sometimes as a pseudo-oscilloscope pattern, or as a meter that indicated phase. Learn how that representation works on your system and watch it while recording in stereo if possible. Or, more critically, during mixing.

Low frequencies are often the most problematic for phase. I once did a project for a record label where the producer wanted to double a synthesized bass part. He thought it would be cool to pan the two parts full left and right. And it did sound amazing, since the changing phase relationship of the bass part made the sound seem to be constantly moving.

I told the producer that it could be a problem on vinyl, but he wanted to use it anyway. I called the mastering engineer ahead of time, to warn him. He must have rolled his eyes, and resigned himself to dealing with the coming challenge.

Another problem arises if there is significant bass content in only one channel. We usually pan the bass to the center of the stereo image, but sometimes other instruments with heavy bass may be panned off-center.

This, too, causes tracking problems for many consumer turntables.

The solution for either of these problems is a trick most lacquer mastering engineers do when necessary. They use a special equalizer that essentially pans the low frequencies into the center regardless of where they were panned in the mix. This process starts at a fairly high bass frequency, like maybe 200Hz, and at decreasing frequencies, the lows get closer the center of the soundstage until they are dead center by 40Hz and below. This is adjustable, so this is just an example.

Our hearing cannot tell the direction most bass sounds are coming from, so putting them all in the center has only a minor effect on the stereo soundstage. It’s the same principle that allows a single sub-woofer to work, no matter where you put it, within reason.

We get directional cues from the overtones of the bass instruments, so the change in localization is not as extreme as you might expect.

The stereo separation we can achieve in the studio can be essentially infinite. But a vinyl disc has only about 20dB of stereo separation. That actually works pretty well, but it is helpful to keep that in mind when mixing, since the extreme isolation between left and right will become much less on the record.

And for that doubled synth bass part? Well, the mastering engineer did what he could do to make the record track properly. The result was a phasey-sounding bass that was not effective at all. It ruined the impact of the record. And back then, vinyl was the only way people heard music, so it resulted in one of the few re-mix sessions I ever had to do back in the analog days.

Now perhaps you see what I mean when I call the vinyl recording process a kluge. Wikipedia has my favorite definition of a kluge: “…a workaround or makeshift solution that is clumsy, inelegant, inefficient, difficult to extend, and hard to maintain.”

So, what can we do in the studio to get the best sound out of a vinyl record?

First, be careful with the high-frequency boost eq you add. The impressive sparkle of the cymbals or sibilance of the vocal may be distorted on the record.

Be aware that the extreme low-end boost that sounds great in your control room is going to reduce the overall level on the record.

If you have input on the sequencing of the songs on an album, remember that the first song on a side will have the best quality, and the last song on a side will have the worst. Any album that has a theme running through it probably has a logical sequence of the songs. But you might not want to put the “hit” song from the album as the last cut on a side.

Another factor in sequencing an album is to make sure that you keep the sides below 22 minutes. If there is a combination of short and long songs, you might have to change the ideal sequence to make the time-per-side work properly. I sometimes have to go to a different sequence for the vinyl version as opposed to the CD version. And since you can put a lot more music on a CD than on a vinyl album, it may be necessary to drop some songs on the LP to make it fit. Or go to a double vinyl album.

Should you make a separate mix for the vinyl version? I don’t, but it may make sense for some albums.

Check with your lacquer mastering engineer ahead of time to find out his or her preferred method of preparing your master for cutting. I find most mastering engineers want each complete side as a separate file, with the desired spacing between each song.

Your mastering engineer needs precise times for the start of each song, its length, and the total length of the side. The space between songs is made obvious on the pressing by a small gap in the groove, made by temporarily increasing the groove pitch significantly. The resulting gap is easy to see. That’s particularly important for broadcast play, if the radio station wants to play from the vinyl album. Make sure you have the desired amount of time between songs with total silence during the gap.

If songs segue from one to the next, the mastering engineer can still delineate the songs, but the music will continue through the increased groove spacing. That works fine, but make sure your mastering guy knows about it.

In the old days, I usually went to the mastering facility at an appointed time to participate in cutting the lacquer master. That way, I could approve the master for pressing, which can be important on a time-sensitive project. A lacquer disc was cut, but on a 12-inch disc, so it could be played on a conventional turntable. We would listen through the entire album to make sure the translation was as good as possible. Then, with all the same parameters, the master would be cut on a 14-inch disc.

I could take the 12-inch reference disc back to my studio to check it one last time, if necessary. You can play a lacquer disc, but each play erodes the disc surface, resulting in increased noise and distortion. Really, only the first play is accurate.

You can still do it that way if you like, as long as the mastering facility is equipped for your attendance. Alternatively, a reference disc can be cut and sent to you for approval. Or the playback of the reference disc can be recorded digitally at the mastering facility. You can approve that recording.

Or you could just trust the mastering facility. In recent years, I have not attended some of the mastering sessions, particularly if they are located far away. I check the recording of the lacquer reference disc.

You may skip this step entirely, if you have confidence in the mastering. I have rarely had to make changes, but sometimes on a demanding project it is good insurance to listen to a reference cut.

One last story about the disc medium that I learned about from the old-timers at the radio station where I worked during high school.

Back before tape recorders became available in the early 1950s, some radio programming was distributed on special discs called “ETs,” for “Electrical Transcription.” These were large 16-inch lacquer discs, cut one at a time. They ran at 78RPM, and could hold about 15 minutes of audio.

An hour-long radio program needed four of these discs. And it was apparent when one disc was segued into the next one because of the abrupt change in audio quality.

The solution was to cut the first disc in the standard outside-in manner, and then cut the next one from the inside-out. Yes, the disc lathe could cut either way. The discs alternated this way through the entire program.

The quality would deteriorate towards the end of the first disc, but that matched the sound of the next disc. No more abrupt change in audio quality.

There are many other details of disc mastering that I have not covered here, including tricks and special effects. I discuss some of them in the episode from 2020 called “Disc Cutting.”

Thanks for listening and commenting. You can reach me at dwfearn@dwfearn.com

This is My Take on Music Recording. I’m Doug Fearn. See you next time.