In one of our previous articles, we talked about the importance of digitizing vinyl records and our role in the preservation of our musical heritage.
Continuing where we left off the last time, we will today introduce you to the ins and outs of digitizing records and the main reasons for doing so.
First of all, digitizing records is a great way of preserving your vinyl. Over time, vinyl records can become scratched or damaged. When this happens, the quality of the sound can suffer. Converting them to digital files will help you preserve the quality of the sound so that you can enjoy them for years to come.
Another reason to convert your vinyl records to digital audio formats is to help you save space. If you have a large collection of vinyl records, and you want to play your music out on location without taking your entire record room along, you can digitize some and take the ones you want to spin with you on the road, right there in your pocket.
Likewise, if you want to travel and enjoy your collection at the same time, you are only limited by the battery power on your phone, laptop or in your car!
Different ways to digitize your music
Below, we will introduce you to the most common way of digitizing your vinyl collection, and show you everything you need to get started.
You’ll often find turntables designed for DJs that look similar to professional models but with an added USB audio interface, making it possible to connect them to a computer and capture the audio directly.
These turntables usually have a price tag of between 110$ and 190$, and usually the capsule and needle are already included. They’re marketed with special packaging and aggressive marketing, clearly aimed at convincing you that they’re ideal for digitizing vinyl. But they are not the most suitable!
Players of this calibre are typically lower in construction quality, as they don’t use proper insulation materials. Furthermore, the audio interface is usually only 16-bit and 44kHz. They typically use direct drive motors, which have less power, and also tend to have 4x more oscillation in playback speed than professional record players.
The rubber disk that comes with most record players and is usually quickly replaced with a velcro on DJ turntables is actually a good ally when you want to start digitizing vinyls. The rubber increases the grip on the turntable, while also absorbing any possible vibrations.
Position the plate correctly and replace the supports
Vibrations can cause havoc on digitizing your vinyl, so make sure to place it on a sturdy and stable surface. If you have a table that is solid and doesn’t shake, that would be perfect.
Handle vinyl correctly for scanning
To take care of your vinyl, hold it by its sides and avoid touching the surface where the grooves are. This way you avoid getting dirt and sweat from your hands on the record.
This is one of a couple of tips for any serious record collector, taken from our article on “How to care of your vinyl records”.
Use a suitable capsule and needle
If you’re thinking about investing in a capsule and needle to digitize and play your records, then I suggest you have two capsules, one for playing and one for digitizing. Hi-Fi capsules can do it well, but we all know how expensive they can be, too.
Digital audio fundamentals
Now that we know the physical elements required to digitize vinyl, we can move on to the intangible: digital audio. A basic understanding of how digital audio works (PCM, or Pulse Code Modulation, specifically) will help you understand both the process and the reasoning behind it. DSD, or Direct Stream Digital, is used for high-resolution music.
When digitizing vinyl, it is common for people to use a 16-bit resolution and a sampling frequency of 44.1 kHz, which is the standard resolution used on CDs. If better means are available, they are worth taking advantage of, as they will result in a better digital copy. To understand why, it is worth knowing two factors that affect the quality of the digitization: the bit depth and sampling frequency. These two factors determine the noise that will be introduced during encoding and the maximum bandwidth that can be recorded.
We begin the process of digitizing audio with an analogue audio signal and take snapshots of it at specific instants of time. These snapshots are called samples, hence the name sample rate, because they express the number of samples we extracted from the audio in one second. For example, 44.1 kHz means that each second of audio consists of 44,100 captured samples.
Some people think that 22 kHz, 22,000 samples per second, is enough to capture any sound a human being can hear. However, this is not accurate. In order to capture sound correctly, the sampling frequency must be more than twice the highest frequency the sound can reach.
An example will help explain this more clearly. If a signal has a frequency of 20 kHz, that means it goes up and down 20,000 times per second. In order to capture both the high pulse and low pulse of the signal, we would need at least two samples per cycle. That means we would need a minimum of 40,000 samples per second.
Why is recording in 48khz important if we cannot hear these frequencies? Though we cannot hear these frequencies, recording in 48khz is important because it captures a higher audio signal. This ultimately provides a better quality recording.
A sinusoidal signal that reaches 15Khz is at the limit of human possibility to be heard. This signal is captured by an interpolar filter that is part of our A/D converters in our audio interface. This interpolar filter fills in with audio signals between samples. The more samples per second, the more the interpolar filter will do a better and better job.
If the bandwidth of an audio signal is restricted to 20kHz, it can be represented digitally in fs (fs = sample per second). At 44.1kHz, filters are more difficulty to implement than when sampling at 48kHz. The “slack” provided by 48kHz provides a more successful reconstruction with fewer secondary defects and the resulting sound will be better.
The bits define the level of detail for which we capture the amplitude, i.e. the “level” of an audio signal. If we capture a signal at 16 bits, we will have a scale of 65,536 points. This value is 2 to the 16 and gives the result 65,536. To measure the amplitude of each of the samples we take every second, we use this scale.
It is important to have high precision when measuring the amplitude of an analog signal. With today’s technology, we can use more bits to increase our precision. For example, if we use 24 bits, we will have 16,777,536 points to measure the amplitude. The more points we have, the more precise our measurement will be.
As we add more definition with higher bit depth, we also increase the music’s dynamic level. And so we will go from the 96dB we can get with 16 bits to the 144dB that 24 bits offer; although these 144dB are theoretical, in reality the converters usually don’t get there. To give you an idea, the Rane MP2015 mixer, which is reputed to have the best converters on the market, offers 116dB at most.
Vinyl’s dynamic range is generally about 60dBs. In comparison, 16-bit audio has a 96dB dynamic range — which isn’t bad. The main advantage of 24-bit audio for digitizing vinyl comes when you’re recording. You can avoid having to align levels, and normalize after recording without worrying about making the noise level noticeable. To be more precise, you can record at 24 bits and then save the file at 16 bits after normalizing the audio. This way you’ll also save space on your hard drive.
To summarize what we have seen so far
If you have a 24-bit audio interface, use its dynamic range when recording, then normalize and save the file in 16 Bits. That way you can save storage space on your hard disk. About the sampling frequency, if your capsule is not capable of reproducing anything above 18khz, then you can choose 44.1 kHz. But if you have a good capsule that goes beyond that, take the 48 kHz one and it will do very well.
Digitizing Music With Audacity
First, configure the software to capture the audio to the desired resolution. Then go to “Preferences” and then the “Quality” tab, and then you will have two options to choose the sampling frequency and a bit depth. In the two menus below, you can choose the real-time conversion type by selecting on Best Quality (Slowest).
Before you start recording, enable real-time monitoring during recording. Go to the preferences in the “Recording” tab and check the option “Playback” through the software (do not choose to playback through hardware, I advise this setting choice because maybe something fails in the software during recording, and you don’t notice it because you are listening directly to the signal coming into the interface, and not to what the software is recording).
Now close the preferences and monitor the incoming signal level. Go to the top left of the meter, and you will see that it has an icon that, when pressed, will display a menu with the option to “Start monitoring”. Enable this option and start playing the disc. Are you not hearing anything, or is the meter not showing any audio signal? Make sure that the interface that the record player is connected to is selected in the main Audacity screen.
Usually at the bottom of the software tool’s panel you have a few options for choosing the input and output device you will use. Once you have an input signal, play the most intense parts of the disc and adjust the input gain so that the signal on Audacity’s VU meter never reaches 0dB. I personally play with this margin a lot, as I record at 24 bits and manage to stay at -4dB. As we have explained before, if you are recording at 24 bits you don’t need to have a signal input level down to 0dBs and try to have a dynamic margin, as you can later normalize the audio without problems.
Once everything is ready to record, make sure that the platter pitch control ( if you are using one that has pitch), is at 0% and place the needle at the beginning of the disc. Click the “Record” button, start the disc, and let it record the entire side of the disc at once.