This page includes supporting information intended to assist in building the stereo preamplifier described in Chapters 9 and 12 of the book, along with commentary from the author. This page picks up where the book left off.
Shown on the right is the second-generation stereo preamplifier described below.
The power amplifiers described in the book include provisions for running the preamplifier using B+ and heater voltages supplied by the amplifier. An industrial-grade connector is used for this purpose. The pin-out of the connectors provides for two separate ground connections for safety.
As an added measure of safety, it is recommended that a separate ground wire (#16 would be a good choice) be connected between the preamplifier chassis and the power amplifier chassis. This ensures that a solid ground is always maintained between the two devices. The simplest method of accomplishing this is to include a ground lug on the back panel of each chassis. The following sequence of components performs well in this application: #6-32 x 1-in screw / #6 lockwasher / chassis / #6 lockwasher / #6 locking nut / #6 flat washer / #6 wing nut.
A ground lug on the preamplifier back panel is also useful as a ground point for input sources, such as a turntable.
It is important to note that the PWB designs developed by the author and available on the Downloads page are intended for tube socket mounting using standoffs as described in Figure 12.2 of Chapter 12 in the book. The PWB will not accommodate other mounting methods. An examination of the PWB design will show that the pads for the socket pins are evenly spaced around the perimeter of the socket. On the sockets themselves, however, the pins are not evenly spaced; there is extra space between pin #1 and the last pin (e.g., pin #7 or pin #9). This was done in order to provide for layout convenience on the PWB. With this approach, the maximum spacing between pads can be maintained, which makes it easier to place traces between socket pins while maintaining generous separation among components. The default layout separation is more than is actually required for the voltages typically present on the PWB. Readers who want to use socket mounting methods other than the one intended for the original PWB design will need to reconfigure the socket mounting pads to match the method chosen. These cautions apply equally to the power amplifier PWBs.
In Chapter 12, the procedure for making initial settings of the input level potentiometers is described. For the Auxiliary and Tuner inputs, an input level of 0.3 V rms is specified. This level works well for many inputs, notably an iPod/iPhone or similar device. Higher input levels may, however, be encountered with some consumer audio gear. For that reason, it may be advisable to set the Auxiliary input for an input reference of 0.3 V rms at 1 kHz, and set the Tuner input level control for an input reference of 1.0 V rms.
As stated in Chapter 9 of the book, the microphone preamplifier is designed for use with a high-impedance dynamic mic. A low-impedance mic will work, but the output level will be low. For those wishing to use a low-impedance dynamic microphone with the preamp, a convenient solution can be found in the Sescom SES-LZHZ-XF14 "XLR Female to 1/4 Inch Male Low Z to High Z Impedance Matching Transformer." This product (there are probably others that perform similar functions) works well with a variety of professional microphones. Tests by the author with a Shure SM58 showed good performance. The SES-LZHZ-XF14 is available from Markertek Online, and probably elsewhere.
While the stereo preamplifier described in the book used RCA jacks for the mic inputs, a 1/4-inch phone jack would be a better choice, as high impedance mics typically use this interface. Low impedance mics generally use an XLR interface, which is provided with the Sescom product. The Neutrix NJ3FP6C-BAG is a good quality 1/4-inch panel-mounted phone jack (Allied stock no. 70088266).
When using a product such as the Sescom matching transformer, take care to keep the device away from magnetic fields, such as power transformers. A magnetic field can produce hum in the device. For that reason, a distance of at least 3-feet is recommended between the power amplifier and the matching transformer. This caution also applies to phonograph pickup elements, which may be subject to the same issues.
An updated version of the stereo preamplifier described in the book has been completed. Changes from the original design include some new features, improved physical layout, and a new circuit board design. The goals were to increase the functionality of the unit and improve (lower) the noise floor. The basic circuit elements remain the same. Extensive use of the first unit has shown the design to meet the stated objectives and to operate reliably.
Specific changes to the stereo preamplifier include the following:
• Improved back panel configuration that simplifies the connection of source devices.
Even with the design improvements detailed here, construction of this unit is challenging. But, the end result is quite nice. Details on this project are given in the following entries.
During the course of building the second-generation preamplifier steps were taken to improve the high-end frequency response of the various inputs. Part of that work focused on the challenge of cable capacitance.
With any cable (two or more conductors, including shield), there exists a distributed capacitance. This capacitance is essentially constant along the cable and therefore increases as the length of the cable increases. A simple experiment was conducted using two common audio cables:
Measurements were made to determine the impact of various volume control settings on the high-frequency response of the circuit, focusing strictly on the effect of the cable itself and the changing resistance of the potentiometer. Taken together, the series resistance of the potentiometer and the shunt capacitance of the cable forms an RC filter.
For the test, a 1 megohm linear potentiometer was connected to a 12-inch sample of each type of audio cable. An input signal was applied across the potentiometer ends, with the shield connected to the low side of the pot (ground on the signal generator). The center conductor of the test cable was connected to the wiper arm. An audio voltmeter (1 megohm input impedance) was then connected to the far end of the cable.
With this arrangement, various settings of series/shunt resistances could be examined. The results, detailed in table format, were revealing. As shown in the table, four settings of the potentiometer were measured. The setting with the greatest deviation from flat response was with 500 kohm series/500 kohm shunt, where response at 20 kHz was down 2.5 dB (relative to 1 kHz). The effect is more pronounced as the frequency is increased.
Note that the table also includes a "Baseline" column. This was essentially a test of the residual shunt capacitance of the audio voltmeter and the minimum measurable capacitance of the bridge. Focusing again on the 500 kohm/500 kohm "worst case" setting, the effect was negligible up to 20 kHz, although surprising large at 50 kHz. One could apply the Baseline readings as a correction factor for the test cable readings, and thereby subtract-out the effects of the instrumentation. This was not done in the table, which simply records the raw numbers.
The data demonstrate the importance of keeping lead lengths as short as possible, and the importance of using low-capacitance cable on high impedance circuits. Also, in order to maintain the same frequency response characteristics between channels in a stereo system, the cable lengths for a given function should be the same between channels.
Based on these experiments, some component value changes are suggested for the preamplifier described in Chapter 9 of the book. Referring to Figure 9.7 and Table 9.10 in the book:
The circuit will certainly work well as described in the book. However, with the lower-value potentiometers, the adverse effects of capacitive loading from interconnecting shielded cable is reduced to a manageable amount; the typical series resistance is reduced by a factor of four for the phonograph and microphone stages, and for the buffer input stage. The changes in value suggested for R18 and R19 reflect the expected source hardware, which will likely be relatively low impedance devices.
As described in Chapter 9 of the book, the minimum load that the phonograph and microphone circuits should see is 220 k ohms. So, the 250 k ohm level controls present no problem for the circuits. The minimum specified load for the tone-control circuit is 100 k ohms; here again, the 250 k ohm Volume control works fine in this application.
Suggested part numbers for the replacement devices are given below:
Another objective of the second-generation preamplifier was to reduce the noise floor. The major noise component of the first generation unit was 60 Hz hum. The minimum noise level of the stereo preamplifier described in Chapter 12 of the book was —60 dB. This is not bad, but not great either. Since the B+ power supply for the preamplifier circuits is taken from the audio power amplifier, in a separate chassis, the assumed source of the hum was the 6.3 V ac heater supply.
To determine the improvement that might be realized with a dc heater supply, the 6.3 V ac from the filament transformer was rectified by a full-wave bridge and filtered by a large-value electrolytic capacitor. The results were quite encouraging, lowering the noise floor by at least 10 dB. Tests conducted on the preamplifier showed the noise floor at —70 dB or lower (better) for all inputs.
The 6.3 V ac filament voltage is applied to a full-wave bridge rectifier (Vishay #6MB05A, Allied stock #70078708). The operating current pull from the preamplifier filaments is about 3.6 A. A 25 A rectifier is specified, however, because the Vishay 6MB05A provides a convenient mounting method (a single mounting screw) and when mounted against the chassis no additional heat sink is required. Filtering is provided by a 4700 microfarad, 50 V electrolytic (Illinois Capacitor #478TTA050M, Allied stock #70112239). When driving the 10 tubes contained in the stereo preamplifier, the output of the circuit is about 6.1 V dc.
Note that with a capacitor-input filter, the output voltage is influenced considerably by the load. Under no-load conditions, the output of a capacitor-input filter will rise to the peak level of the input ac waveform (1.4 times the rms level). Therefore, if a dc filament supply is used, do not operate the preamplifier with multiple tubes removed.
The impact of a dc filament supply on the life expectancy of the tubes used in this design, if any, will be hard to assess. Nonetheless, the author will keep an eye on this going forward. (Note: as of January 2017, four years after making this change, no tube failures have occurred.)
With the rectifier mounted against the chassis, the modification is easy to implement. No changes are made to the PWBs, and the Hum Balance control performs the same function as with ac on the filaments. The setting of the Hum Balance control has a surprisingly large impact on the noise floor when using a dc supply.
Readers may recall that in Chapter 12, I mentioned that shields were not used over the phonograph and microphone preamplifier tubes. At the time I wondered how much of an improvement in the noise floor I might gain by adding shields. During final tests on the second generation stereo preamplifier I tried adding shields to the 7025 phono preamp and 5879 mic preamp tubes. With the dc heater addition there was no observable difference in the noise floor with or without tube shields. In all cases, the noise floor was well below —70 dB, relative to 1 V rms output from the Line Out jacks. Predictably, there was even less noise when measured from the Tape Out jacks, since the tone-control stage is bypassed when the Tape Out jacks are used.
Still, after a short search, I came upon some very stylish tube shields for 9-pin miniature sockets. These devices are bolt-on compatible with the Belton sockets specified in the parts lists contained in the book. The shields are available in a variety of colors, including yellow, red, green, blue, and black. Check them out on the Antique Electronic Supply Web site. The shields cost about $1.50 each and look very nice. I have retrofitted the two preamps that I have built with these shields on the phono and mic input stages.
The cost of building any project is an important consideration for the audio enthusiast and hobbyist. The focus of the first-generation stereo preamplifier was to optimize the various elements that went into the project. Cost was a secondary consideration. For the second-generation unit, an effort was made to accurately determine the bill of material (BOM) costs and the time needed to build the preamplifier.
The total BOM (including shipping) for the second-generation stereo preamplifier was $1,570. The major cost centers included:
Builders could eliminate the PWB cost by using hand-wired terminal strip construction. Other potential areas of cost reductions include the back and top-side decals, and the Plexiglas cover.
The total time required to build the second-generation stereo preamplifier—from ordering parts to completing performance measurements—was approximately 22 hours.
Note 10 (January 2017)
Detailed step-by-step documentation is provided for the second generation preamplifier in a 108 page "User and Assembly Manual" that is available for download. If you are familiar with the Heathkit assembly manuals of the past, the approach taken with this manual should seem comfortable. For those who want a printed version, it is available for purchase on Lulu—a web site for specialized, print-on-demand documents. (There is charge of $14.95 for the printed manual; see "Stereo Preamplifier: User and Assembly Manual" on the Lulu web site.)
For readers who have downloaded the Acrobat file or purchased the printed manual, a ZIP file is available for download that will assist in building this preamplifier. The ZIP file contains the following individual files:
• Schematic diagram of the preamplifier as an Acrobat (".pdf") file
Note that the files above are provided as-is. Every effort has been made to make sure they are complete and accurate, but no warranties are expressed or implied. Builders are encouraged to double-check the information contained in the above files prior to proceeding. For the front panel layout, users can customize the text as desired; e.g., "Built by John for Mary, January 2017." All files are provided for personal use only. No further distribution is allowed.