The technology of music recording has a long and interesting history—from cylinders to vinyl to CDs and beyond. In this article, Jeff discusses those historical steps. He then shares the details of a project that connects a record player to a modern Bluetooth music player. Jeff figures out how to provide both filtering and amplification of the turntable cartridge’s signal by designing a preamplifier with some filtering characteristics.
Have you ever heard Thomas Edison recite “Mary Had a Little Lamb?” The audio is from an interview with him about his recording device. The prototype recorded the audio vibrations shouted into a megaphone onto a tin-foil-wrapped cylinder. A shop lathe was used to cut threads into a metal shank rotating in sync with it. This was the basic mechanics of Edison’s recording machine (Figure 1). As the cylinder was hand cranked, audio vibrations caused the cutting “needle” to stray from its normal path, depositing irregularities in the groove’s width. When the needle was relocated to the beginning of the groove and the cylinder again cranked, the needle followed the original groove, and this time the recorded variations caused the needle to reproduce identical vibrations, thus playing back the original audio. The fragile foil had a very limited playback lifetime, however.
After material improvements allowed the recorded cylinder to be copied through the molding process, the master recording mold could be to be used to mass-produce copies, and a new industry was born. By the end of the 1800s, the cylinders had been squashed flat, but still retained a single groove that spiraled inward. This simplified the playback mechanism. The record could be produced with a different audio track on each side! However, without the electronic amplification that we take for granted today, groove vibrations were still being transferred to the air by an exponentially shaped horn that mechanically amplified the sound. No boom boxes here!
The 1920s brought the use of microphones and vacuum tube amplifiers. This produced better recording on the master side, and enhanced listening on the playback side. As the fidelity improved, it was seen that the lower frequencies required larger groove excursions (wider grooves). This wasn’t an issue if these lower frequencies were reduced. This, of course, led to an imbalance in playback volume. It was decided that if a high-pass filter was used on the recording side, then a low-pass filter would cancel this imbalance for playback. This was established as a standard by the Recording Industry Association of America (RIAA) around the 1950s.
The first 10” records—or “platters,” as they were called—required a rotational speed of 78RPM, and held 3 minutes of audio. Most of them were made of shellac and were very brittle. Material and technical improvements allowed for the micro groove and slower speeds, which lengthened the total recording time per side. The single or “45” became popular as the growing music industry found a captive audience demanding access to an explosion of recording artists. At 7″ and 45RPM, these had a comparable recording time to that of the older and fragile “78’s.” The use of vinyl created a more durable and inexpensive product. A larger, 12″ album or LP (long play)—which ran at 33 1/3 RPM—touted up to 30 minutes of audio per side. This became the staple of the industry, and featured something new beyond a collection of multiple music tracks—album art (Figure 2).
The recording industry is always looking to take advantage of the newest technologies. In an effort to give the user a better experience, the 8-track and cassette-magnetic-tape formats attempted to take over the record market. The convenience of taking your music with you via car players and the Sony Walkman really changed the market. Tapes eventually gave way to the compact disc (CD). Today, most of our audio playback devices use solid-state media storage. However, this move from true analog recording to a digital representation has created quite the controversy.
Audiophiles are people who are enthusiastic about high-fidelity sound reproduction. They seek to reproduce the sound of a live musical performance, typically in a room with good acoustics. It is widely agreed that reaching this goal is difficult, and that even the best-regarded recording and playback systems rarely—if ever—achieve it. That being said, most audiophiles still believe that platters reproduce recorded sound more accurately than their digitized counterparts.
Today’s media centers are designed more for video than audio. Soundbars give great playback for theater quality sound that enhances our home theater experience. Many soundbars are Bluetooth, and connect wirelessly with the signal’s source.
This month’s project will bind platter players to these new sound systems. It will allow you to connect your turntable to your soundbar by adding a preamp and Bluetooth transmitter. As noted earlier, an RIAA filter is used when recording audio that will be pressed into vinyl, and the opposite filter must be added to the playback loop to compensate. Let’s start with the playback stylus to see what we will need.
A stylus (needle) is the part of the cartridge on a record turntable that interfaces with a record’s surface and tracks the modulations (deformities) in a groove. There are hundreds of stylus manufacturers, but years ago, I chose Audio Technica for my turntable. They make a broad range of phono cartridges that cost from $29 to thou$and$!
Figure 3 shows the parts of a moving-coil cartridge. The cartridge is the most expensive component of a turntable, because it must interpret the groove imperfections to replicate the original audio as closely as possible. It is located at the end of a pivoting tone arm, and when placed at the outside edge of a rotating record, it must be heavy enough to be dragged by the groove along the groove’s spiral path, inward toward the record’s center. The groove has been deformed or modulated by the recorded audio.
The cartridge also must be light enough to render the grooves imperfections accurately without damaging them. These imperfections vibrate the stylus and the associated magnets. Nearby coils detect the magnet’s motion and produce an electrical signal, which follows the groove’s imperfections. The cartridge output signal is extremely small, and must be amplified to match the approximately 1V peak-to-peak (VP-P) input expected by an audio (power) amplifier.
THE RIAA FILTER
Here is where the RIAA filter comes into play. We can provide both filtering and amplification of the cartridge’s signal by designing a preamplifier with some filtering characteristics. I found an app note from Maxim Integrated (Maxim App Note #1931)  for a USB-powered, high-precision, Hi-Fi phono amplifier with RIAA equalization, employing a Maxim Integrated MAX4478 quad op amp.
This wide-band, low-noise, low-distortion quad op amp offers rail-to-rail outputs and single-supply operation. This app note was written by Aussie Rod Elliott. The design response curve in Figure 4 is from his website . Note Pole #1 initiates a roll-off of 6dB/octave at 50Hz, while Zero #1 at 500Hz cancels it. Pole #2 again initiates additional roll-off at about 2,100Hz. The end result will be a boost (20dB) of the low frequencies less than 500Hz, and a cut (-20dB) in high frequencies greater than 2,100Hz.
The schematic for this preamp is given in Figure 5. While I don’t have an adequate sweep generator and bode plotter to do a precise analysis of the circuit, I used an audio oscillator and scope to manually measure a few points from 20Hz to 20KHz. I found the low-end boost (less than 200Hz) and high-end attenuation (greater than 2,000Hz) to be as advertised.
With the required signal massage, the audio can now be routed through the PCB’s 1/8″ phone jack to an amplifier. Or, with the addition of the KCX_BT_Emitter, it can be digitized and sent through the USB connection to a PC. The PC will recognize the USB input as an audio device. However, I want to send it wirelessly to any Bluetooth receiver, like a set of wireless speakers that can be located anywhere within range of the turntable transmitter.
KCX AT COMMANDS
As with many imported products, I found little or no information available about the KCX_BT_Emitter or the chip used, on the ZhuHai JieLi Technology, Ltd. company website. Suggested uses are implied by a few connection diagrams, as serial communication device, a USB audio device or wirelessly via Bluetooth. What to do once you’ve wired a circuit is left up to the purchaser’s imagination.
Apparently, this module has the ability to interact with a user through AT commands. I just happened to stumble on this on the Electro-Tech-Online.com forum (search for KCX) . I can’t understand why a manufacturer would sell a product with so little documentation. Anyway, according to “DrG”‘ on the forum, connect a serial dongle up to TX, RX and ground and you can talk to the module at 9600 baud. Type in AT+ (no CR or LF), and it should respond with OK+. When powered, it will attempt to connect with any device in its list, starting from the first. The list or lists can be by device MAC address or Name. If no devices in the list are available, then the KCX will begin scanning for devices. In Listing 1, you can see the results if connected via TX/RX on your terminal. I’ve tabbed the replies for clarity.
LISTING 1 - Type in AT+ (no CR or LF) and it should respond with OK+. When powered, it will attempt to connect with any device in its list, starting from the first. Here, you can see the results when connected via TX/RX on your terminal. I've indented the replies for clarity. AT+REST OK+REST POWER ON AT+SCANOK+SCAN New Devices:1,MacAdd:0x4cbb58bab6d5,Name:JEFF10 ALL Devices=1 MacAddr=0x4cbb58bab6d5,Name=JEFF10 New Devices:1,MacAdd:0x169bd938586c,Name:FOUR40W-SEVIZ New Devices:2,MacAdd:0x4c11ae70f3aa,Name:ESP32 SXT Monitor ALL Devices=2 AT+ADDLINKADD=0x4c11ae70f3aa OK+ AT+ADDLINKNAME=FOUR40W-SEVIZOK+ADDLINKNAME VM_Name 1 =FOUR40W-SEVIZ New Devices:1,MacAdd:0x169bd938586c,Name:FOUR40W-SEVIZ CONNECTED (device turnned off) DISCONNECT (device turnned on) ALL Devices=2 MacAddr=0x169bd938586c,Name=FOUR40W-SEVIZ MacAddr=0x4cbb58bab6d5,Name=JEFF10 New Devices:1,MacAdd:0x169bd938586c,Name:FOUR40W-SEVIZ CONNECTED AT+VMLINK? OK+VMLINK BT_ADD_NUM=2 BT_NAME_NUM=2 Last_Add=0x169bd938586c VM_MacAdd0=0xbff0f09a7839 VM_MacAdd1=0x4c11ae70f3aa VM_Name0=OontZ Angle VM_Name1=FOUR40W-SEVIZ (device turnned off) DISCONNECT (KCX unpowered) (KCX powered) POWER ON New Devices:1,MacAdd:0x169bd938586c,Name:FOUR40W-SEVIZ CONNECTED AT+VMLINK? OK+VMLINK BT_ADD_NUM=2 BT_NAME_NUM=2 Last_Add=0x4cbb58bab6d5 VM_MacAdd0=0x4c11ae70f3aa VM_MacAdd1=0xbff0f09a7839 VM_Name0=OontZ Angle VM_Name1=FOUR40W-SEVIZ ALL Devices=0 (KCX button pushed) Delete_Vmlink ALL Devices=0 AT+VMLINK? OK+VMLINK BT_ADD_NUM=0 BT_NAME_NUM=0 Last_Add=0x00000000000 (device turnned on) New Devices:1,MacAdd:0x169bd938586c,Name:FOUR40W-SEVIZ CONNECTED AT+VMLINK? OK+VMLINK BT_ADD_NUM=0 BT_NAME_NUM=0 Last_Add=0x169bd938586c
You’ll note that if the KCX is not connected, it will begin scanning for available devices. On the first scan, it found my computer Bluetooth (JEFF10). On a second scan, it found the Bluetooth speakers (FOUR40W-SEVIZ) and an ESP project on my bench (ESP SXT Monitor). I added the speakers in both formats ADDLINKNAME and ADDLINKADD. After the reply, I immediately received CONNECTED. When I physically turned them off, I got a DISCONNECT.
The AT+VMLINK? command shows the devices I had added to the list. I had formerly added another set of speakers manually, because they were not being seen and would not connect. (Note: “OontZ” (Bluetooth speakers) doesn’t show up in any of the scans.) If I powered the KCX (project PCB) off and back on, I got a connection to the FOUR40W-SEVIZ speakers. The AT+VMLINK? command shows all devices are still in the list. When I pushed the button on the KCX, I got a Delete_Vmlink message telling me it had erased all the stored data (room for 10 entries). The AT+VMLINK? command shows no devices are in the list.
As soon as the speakers were turned on, a connection was made. Okay, the wireless Bluetooth seems to works as intended, so let’s wrap this up.
Every project seed is planted when some interesting technology is discovered. In this case, when I ran across the KCX_BT_Emitter, it was being advertised as an audio transmitter. This was different from all the Bluetooth modules I’d seen in the past. And the price was certainly appealing. I happened to be storing away some project boxes in an attempt to unclutter my office space, when I came upon containers holding hundreds of record albums.
Every time the entertainment industry moves to a new technology, they open the flood gates of users clamoring for their favorite music, movie or other recreational content on the new medium. The digital world brought on anti-pirating schemes preventing duplication of copyrighted material. I certainly believe that artists should be compensated for their material. That said, I also believe that once you’ve paid for material, you should have the right to it on any media. Therefore, transferring it to another type of storage for your own use should be your prerogative.
There were evenings in the past when I’d do nothing but listen to music albums. With the onslaught of big screen TVs, family entertainment has shifted focus. I no longer have a stereo system—turntable, power amplifier, equalizer, tuner and so on—set up in the den. I have the room for a turntable only if I don’t need all the other peripheral equipment to play an album. With that in mind, this project came about to satisfy that requirement. The project circuitry would need to fit inside the turntable—and have no additional connections necessary.
The first step was to create a schematic. This would include all the previously discussed circuitry, RIAA preamp, KCX_BT_Emitter, voltage regulation and a few connectors for the turntable’s cartridge, USB and optional preamp output. I used a eared enclosure, the WM-011 from Serpac Electronic Enclosures. The mounting tabs or ears allow it to be easily attached to the inside of the turntable’s base. Taking measurements from the manufacturer’s drawings, I created an outline for the PCB that would fit into the enclosure.
Because the main amplifier was only available in a surface mount (SMT) part, I decided to use SMT parts where applicable. Hand-wired prototypes usually use through-hole parts, so this one would need a PCB. Designing with through-hole parts is always easier and quicker, since you can make modifications to your wiring simply by unsoldering a wire. Although you may be able to change a part’s value on a PCB, changing the wiring is much more difficult and usually leads to another round PCB design. Locating SMT parts is always a challenge. You may be required to use a different-sized part just because of availability (or the lack of). For this reason, it is always a good idea to get your parts ordered and on hand before laying out the PCB.
I am used to going right from schematic to layout. This creates issues when you can’t find the part size you thought you could get, as I just mentioned. So, if you can’t wait, you may need to get creative to get a part to fit in the designated outline. You must remember that you most likely will be hand soldering parts onto the PCB. This means that you not only must be able to pick up and place components, but also you must have soldering equipment made for attaching tiny parts. I have successfully soldered SMT parts with the 0.1″ tip of a standard iron, but you will be much more productive using a tip size that is no greater than the width of the IC leg that you are soldering.
For years, 0.31″ diameter rosin core tin/lead solder has been the defacto standard for hand soldering parts. The push to eliminate lead from any part or solder has been a rough road. It is difficult to solder an SMT part using solder that’s thicker than the part or the iron’s tip. Solder pastes make this easier. A solder paste is a fluid flux that contains microscopic solder balls. A small dab placed on each solder pad will hold a part in place, and also will attach it when a heat source is applied. I generally paste only opposite corner pins of an IC, so I can see the pin/pad alignment of the other pins and get it aligned correctly before attachment. Any misalignment can be changed easily by alternately reheating and adjusting the corner pins before soldering the entire chip.
Although SMT circuitry requires a steady hand, ample magnification and plenty of patience, don’t pass it up if you have the chance—it’s actually good therapy. You’ll note that the Bluetooth module uses 0.05″ pin spacing. I found a matching pin header, but needed to order one with more pins then necessary (stocking issues) and carefully clipped off the excess. Using a socket for any modules allows it to be detached for any reason. I always begin debugging without these attached, to prevent wiring error catastrophes.
The circuit is powered by 5V. Since a USB connection can supply this, I used a USB charging cube with a USB connector. These are plentiful, and you might even have extras hanging around. I am only using the USB connection to power the project. Its AC plug can be wired into the turntable’s AC connection, or you can use a small multiple-outlet extension cord, and plug-in both the turntable and USB charger beneath the turntable. Figure 6 shows the finished preamp/filter circuit PCB with the KCX_BT_Emitter plugged into a vertical header. Powered by a USB charger cube, the Bluetooth module will attempt to connect with its last “paired” receiver and stream audio wirelessly.
I might change the location of the LED and push button, so they are accessible through the bottom of the enclosure. This will require a couple of additional holes for mounting to the inside of the turntable. That way, they will be accessible from the exterior of the turntable. I can already think of several other uses for the KCX module, but for now I’ll be enjoying a resurgence of listening to classic albums—especially with most network television now into reruns. Too much to do, too little time.
PUBLISHED IN CIRCUIT CELLAR MAGAZINE • SEPTEMBER 2020 #362 – Get a PDF of the issue