MC 2255

STEREO POWER AMPLIFIER

McIntosh

The Compendium

FRONT PANEL: Glass with illuminated teal/gold nomenclature. Illuminated left and right power level meters (linear WATTS or DECIBELS). METER function lamps: WATTS, HOLD or DECIBELS. SYSTEM TEST digital readout. POWER GUARD lamps: NORMAL or LIMIT. 7 LEFT GAIN control. METER RANGE SWITCH: -20, -10, 0, WATTS and HOLD. HEADPHONES jack. SPEAKERS switch: OFF or ON. RIGHT/MONO GAIN control. POWER switch: OFF or ON. PANLOC buttons.
BACK PANEL: LEFT and RIGHT speaker OUTPUT barrier strips. Power cord. AC OUTLET. FUSE. MODE switch: STEREO, MONO BRIDGE or MONO PARALLEL. LEFT and RIGHT (MONO) audio INPUT. INPUT LEVEL switch: 2.5V or.75V. Panloc mounting. SPECIFICATIONS: 250w/ch. (500w mono). Response 20-20kHz (+0 -0.25dB). Distortion 0.02%. Noise and hum -95dB. Output impedance 1, 2, 4 and 8. (0.5, 1, 2, 4, 8 and 16 ohms in mono). Damping factor 30 or greater. Input impedance 50k. Input sensitivity 0.75v or 2.5v. Headphone impedance: low. Sentry Monitor. Power Guard. 2007 VintageMac Larry Hodson Page 1 / 2

Rev.107

MC 2255 - Sold from April 1981 - June 1986. S/N: CT1001~CT5700 (4,699 UNITS) Size 7-1/8"H, 16-3/16"W and 14-1/2"D. Weight 82 lb. 2007 VintageMac Larry Hodson $2,995.00 Page 2 / 2

THE MCINTOSH POWER METERS
Mclntosh developed output monitoring meters add to your operating flexibility. Ordinary meters are incapable of indicating the short interval information in a sound wave. The mass of the meter movement is too great to respond to the instantaneous changes in music program material. That short interval information can have a duration as brief as onehalf of one thousandth of a second. Even should the meter be capable of the high velocity movement the human eye could not perceive the information.
Oscillogram of output waveform with and without Power Guard. Input overdriven for each trace 20 dB.
Mclntosh engineering pursued both problems electrically by developing new electronic circuits that cause the meters to respond to short interval information with an accuracy of 98%! To permit the eye to see such high speed motion the electronic circuits that drive the meter pointer are time stretched so the meter pointer position can register in the persistence of vision characteristics of the human eye. The meters indicate directly in watts, or can be made to hold the highest reading and continuously update on higher power or can be switched to be peak reading peak locking decibel meters.
In only a fraction of a millisecond Power Guard dynamically reduces input level to prevent amplifier overload yet permits the amplifier to deliver its absolute maximum power output without extra distortion. In addition, the output of the "waveform comparator" activates the front panel NORMAL and LIMIT indicators. The Power Guard circuit provides a precise visual indication when the amplifier has reached full power output. Any time that the input circuit is fed excessive amounts of signal causing waveform differences through the amplifier of 0.5%, the output mode indicators change from green NORMAL to red LIMIT automatically and instantaneously. This warning persists long enough for positive visual indication of clipping for a pulse that is so infrequent or short that it would be impossible to be seen even on an oscilloscope. The indicators will illuminate on clipping for a pulse as short in time as 100 microseconds. You are always assured that the power of your amplifier is as clean and distortion free as it can be.
When used as a watt indicating meter all the information is direct reading, without conversions or complicated mathematics. In addition, as direct reading meters they are calibrated in average watts for a sine wave signal but respond to signal peaks. The meters indicate direct power in watts. They are calibrated in average watts for a sine wave signal but respond to signal peaks. So, a 200 average watt indication also means 400 instantaneous peak watts. The meters are voltage actuated and indicate power accurately when the amplifier is operated into rated output load impedances. Watts Hold, permits the meter to lock to and indicate the highest power peak in a sequence of peaks. The meter will be driven to maximum power and electronically held there until a higher peak passes through the amplifier. If no further peaks are reached the meter needle will very slowly return to a lower peak or to its rest position at a decay rate of 10 dB per minute.

The meters have extremely fast rise time, about 500 microseconds for 90% accuracy. A tone burst of 500 microseconds is almost inaudible even at full power. The meters are protected from damage in the event of overloading in the wrong meter range.

AUTOMATIC TEST SYSTEM

THE MCINTOSH OUTPUT CIRCUIT DESIGN
To achieve long trouble free life in an amplifier it is essential to have cool operation. Cool operation results from the careful design of the output circuit, matching of the output circuit to the loudspeakers with an autotransformer and a mechanical design that permits the use of generous sized heat sinks providing adequate ventilation without the use of fans. The bipolar eptaxial output transistors and the Mclntosh output circuit allows the amplifier to operate as cool as possible. When there is limited program demand on the amplifier only the optimum number of output devices operate. When there is no signal no output device is conducting. Conservative Mclntosh engineering keeps operating temperatures low assuring long life. The interleaved multifilar wound Mclntosh designed autotransformer transfers all the power you paid for to all impedance taps. You are not power penalized for operating at an output impedance of less than 8 ohms. The Mclntosh autotransformer does its outstanding job without adding phase shift, limiting frequency response or power output. The distortion through the autoformer is 0.003% at 20 Hz and unmeasurable at higher frequencies. In short, the Mclntosh autotransformer is the ideal answer to a difficult problem. Heat sinks must be large and they must have adequate ventilation for effective cooling. For example the MC 2255 has 1100 square inches (7.64 square feet) of radiating surface. In addition, the chassis has been designed to permit the maximum amount of air to flow over the heat sinks to conduct away the life limiting heat. Mclntosh amplifiers provide the correct connection impedance to drive numbers of speakers simultaneously. For instance the 1 ohm output will drive eight 8 ohm speakers and deliver full power without overheating.
The Automatic Test System provides positive protection and extends the long trouble free life of an amplifier. Each time the amplifier is turned on an Automatic Test System measures and verifies accurate performance at seven critical points in the amplifier's circuits. The Automatic Test System verification assures operational readiness before operation starts and limits any damage should there be component malfunction. Each time a test is verified an LED number indicator shows which test is being performed. An adjustable "beep" tone can be heard with each test.

(* - - - Continuous operation not possible due to overheating. Protection circuit is assumed to current limit when load falls below 4 ohms, in actuality the o u t p u t into 4 ohms and lower impedances will fall below the values shown.)
Fig. 7 - Performance of non McIntosh transformerless amplifier rated for 300 watts in 8 ohm load
Fig. 5 - McIntosh MC 2205 o u t p u t stage and transformer
dred to one. With transistor output stages, a ratio of only about 4 to 1 is required. In tube amplifiers, extremely good balance in the push-pull primary was required if notch distortion was to be avoided. Now, using a single ended push-pull transistor output stage the transformer can be driven in a single ended fashion. One end of the winding Is returned to ground potential. With the transformer at ground, no isolation is required between the input and output and therefore a simple autotransformer can be used. Fig. 5 shows a typical arrangement used in our new MC 2205 amplifier. The output stages are designed to work o p t i m a l l y i n t o a load impedance of 2.1 ohms and it becomes a simple matter to "tap into" the auto-transformer for that precise impedance match. Taps for 1 ohm, 2 ohm, 4 ohm and 8 ohm operation are arranged so that the output transistors continue to work into their optimum impedance. The result: full power output at any of these impedances, with no possibility of thermal over-dissipation. Our popular MC 2300 amplifier also uses an auto-transformer and Fig. 6 shows how that amplifier is able to deliver its full rated power (300 watts RMS per channel) into any impedance from 0.5 ohms to 16 ohms, as well as to 25 volt and 70 volt multispeaker system taps on the transformer. If we compare these results with those obtained with a similarly rated
transformers in a moment) can introduce about 3 degrees of phase shift at 20 kHz (Fig. 8A), which is certainly insignificant. The typical volume control used on amplifiers (both those that are OTL and those equipped with transformers) introduce more s h i f t t h a n that about 20 degrees in fact (Fig. 8B). Since an output transformer is driven from an extremely low impedance, there is actually more lowfrequency phase shift caused by the usual input coupling capacitor at these low frequencies than by the transformer. So, why haven't more manufacturers used output transformers on solidstate amplifiers? Possibly they are not aware of the technology, but more likely they don't want to spend the extra cost. A good transformer is an expensive component. It is heavy, takes up a fair amount of space and contradicts the audiophile's notion that transistorized equipment must be small and lightweight. Be that as it may, the FTC regulations suggest that o u t p u t transformers are the only logical solution to rating audio amplifiers honestly at 4, 8, 16 or any other impedance required.

Fig. 8A - Typical of phase shift in McIntosh auto-transformer at 8 ohms
OTL amplifier (Fig. 7) we see that at all but 8 and 16 ohms, continuous operation at theoretical maximum power is impossible because of overheating and protection circuit limiting. Operation at 16 ohms, though possible, is limited to a maximum power output of 150 watts, in this case, while operation into a 70-volt line is impossible because of limitations in power supply voltages. What About Phase Shift? Critics opposed to the use of transformers in output circuits of audio amplifiers arc quick to point out that "transformers introduce phase shift" at the low and high frequency extremes. As a matter of fact, a properly designed transformer (and we'll get into some of the factors that are involved in designing McIntosh output
1 watt, 8 ohms Volume Control Clockwise Volume Control 12 o'clock Fig. 8B - Typical phase shift in a complete Mclntosh amplifier
Not Just Any Transformer! At M c l n t o s h , we wind all our own output auto-transformers. Of course, we could purchase them from any one of a number of transformer companies who do nothing but wind transformers (our power transformers are, in fact, purchased from other suppliers), but we have long since found that transformers can't always be made successfully "according to the book". A great deal of experimentation is required before a new design of a transformer can
be mated to a specific amplifier circuit. We went through dozens of developmental samples in the case of our new MC 2205 amplifier. What we ended up with it shown schematically in the diagram of Fig. 9. The transformer is trifilar wound to provide coupling between sections. It takes 23 individual windings to make this output transformer. There are five different winding sections, all five of which are connected in parallel. We use grain oriented silicon steel core laminations because that kind of core means less iron-and less iron in turn means tighter coupling. It also means lower winding resistance for a given size wire. The grain oriented silicon steel means that it has a higher magnetic saturation point-about 17 kilogauss versus 12 to 13 kilogausses for the non-oriented variety. There is therefore less core loss, or, to put it simply, we end up with a more efficient transformer-one which couples more of the available amplifier power to the speaker loads. To further improve coupling, we don't use any interlayer insulating paper in a power transformer that might pose a breakdown problem. But since our polyurethane insulated

TAPE MONITOR: Two pushbutton switches. Either of two tape recorders may be monitored. TAPE COPY SWITCH: Two pushbutton switches. Either of two tape recorders may be connected to copy from one to the other in either direction. LF FILTER (Rumble Filter): Flat or roll-off 6 dB per octave below 50 Hz, down 12 dB at 20 Hz. HF FILTER (Scratch Filter): Flat or roll-off 6 dB per octave above 5kHz, down 12 dB at 20k Hz.
C 504 is a compact state-of-the-art preamplifier.
PERFORMANCE LIMITS and RATINGS C 504 is the same as the preamplifier section of the MX 117 found on page 38. MECHANICAL INFORMATION SIZE: Front panel measures 16 inches wide (40.6 cm) by 3 5/8 inches high (9.2 cm). Chassis measures 14 3/4 inches wide (37.5 cm) by 2 3/8 inches high (6.0 cm) by 14 1/2 inches deep (36.8 cm), including connectors. Knob clearance required is 1 1/4 inches (3.2 cm) in front of mounting panel. FINISH: Front panel is anodized gold and black with special gold/teal nomenclature illumination. Chassis is black. MOUNTING: Mclntosh developed professional PANLOC. WEIGHT: 14 pounds (6.4 kg) net, 25 pounds (11. kg) in shipping carton.

Electronic Laboratories'

PRODUCT ANALYSIS
THE McINTOSH C 27 STEREO PREAMPLIFIER
manufacturers. Well, for all those thousands of dedicated Mac fans let us preface this report by stating emphatically that the newly designed C 27 is much more than a cosmetic face-lift of older models. In fact, the highly acclaimed "Mclntosh look" of the front panel has been retained, while new internal circuitry, from stem to stern, makes this new entry a state-ofthe-art design that should enjoy the same sort of longevity typical of other Mclntosh products. The earlierintroduced C 32 was, as we said, a multiple featured product, with such built-in accessory features as a full five-control graphic equalizer and an adjustable linear expander, not to mention its fully separate monitoring/headphone amplifier output facilities. The C 27 takes a more restrained approach in which those features that are deemed important to most preamplifier/control users are retained but other accessory facilities are omitted in favor of a suggested retail price that will be affordable to a greater number of potential purchasers. Think of the C 27 as a "middle of the road" unit in terms of its control facilities and flexibility. It is not a unit that subscribes to the "straight wire with gain" approach - an approach which we, at least, have always felt is a bit unrealistic. After all, how many of us are blessed with companion components, speakers, or even room acoustics that require absolutely no control "tweaking"? Rather than do away with such control features as tone controls or filters, it seems to us that it is far more useful for the consumer to have these facilities providing that their presence in the signal path does not introduce any form of distortion. Evidently Mclntosh felt the same way and, as we learned during our lab testing of the C 27, these features, when deactivated or, in the case of the tone controls, when placed in their nominal flat positions, do not in any way negate the "straight wire with gain" concept which some purists espouse. The front panel of the C 27, pictured in Fig. 1, is anodized gold and black with Mclntosh's familiar goldteal panel nomenclature illumination which magically appears when power is applied. Rotary controls along the top of the panel include a five-position program input selector (aux 1, aux 2, tuner, phono 1 and phono 2 ) , a mode selector (left or right to both outputs, stereo reverse, stereo, L + R mono, left plus right to left or right outputs) and, at the extreme right, the master volume control. The four rotary control sets at the lower right of the panel are dual concentric types. The first of these introduces, singly or in combination, the high and low cut filters and selects either or both sets of loudspeakers which can be switched via the front panel providing that the power amplifier's outputs are connected to the rear of the C 27 instead of directly to the speakers. Bass and treble tone controls permit tonal tailoring of each channel separately while the rightmost control in this group is a combination balance and loudness control. The loudness control of the C 27

Fig. 5 - Distortion versus frequency, for rated output (2.0 volts) aux input to main outputs
under 0.04%. Maximum output obtained from the main output terminals was 0 volts for a rated distortion figure of 0.05% harmonic. Tone control action of both the bass and treble controls is so precise and well calibrated that we decided to trace overall response for each of the ten possible settings of each of those controls. The results are displayed in the 'scope photo of Fig. 6. Flat response is re-
Fig. 4 - Playback response of phono preamp-equalizer section
though our frequency sweep extends only down to 20 Hz, the response continues to roll off below that frequency at a 6 dB per octave rate, exactly as prescribed by the new IEC standards. If you try to measure the phono response against the "old" RIAA curve, however, you will find that at 30 Hz it differs from the old prescribed point-by-point plot by approximately 1.3 dB In terms of the new and preferred playback curve, the equalization of the C 27 was so accurate that it varied by no more than 0.1 dB at any of the remaining test points measured. Fig. 5 is a plot of distortion versus frequency (at rated output) via the high-level inputs, as observed at the main output terminals. Over the entire spectrum from 20 Hz to 20 kHz, harmonic distortion never exceeded 0.025% and at mid frequencies it measured an extremely low 0.009%. Since the distortion measured from the phono inputs to the tape outputs was even lower, we did not bother to plot these results since the high-level section would govern actual distortion of the output signals under actual use condition. Though Fig. 6 - Step-by-step tone control response characteristics
presented by the center curve of this series and was measured as flat within 0.5 dB from Hz to 23 kHz. The -3 dB points in response occurred at 4 Hz and 60 kHz. Signal to noise ratio in the high level settings measured 86 dB below rated output, unweighted, and
92 dB with an "A" weighting network inserted, both figures exceeding published limits of Mclntosh. Volume control tracking was found to be accurate from left to right channels within 0.2 dB from maximum settings down to a -70 dB level The action of the high and low cut filters is illustrated in the 'scope photos of Fig 7 While the slope rates are clearly only 6 dB per octave (we prefer steeper slopes ourselves), Mclntosh wisely set the c u t - o f f points at sufficiently high and low frequencies ( Hz for the low-cut and 5 kHz for the high-cut) so that their intro-
shown in Fig. 8. Note that these curves were achieved with a constant setting of the master volume control and only the loudness control setting itself was varied, from one extreme to the other. Total level variation a f forded by this control amounted to approximately 22 dB (each vertical division in this Figure as well as in Figs. 4, 6 and 7 represents an amplitude difference of 10dB). Use and Listening Tests As is true of other Mclntosh components we have tested in the past, one immediately senses that the C 27 is, first and foremost, a reliably built product and one that is likely to last for a very long time and require a minimum of servicing. Controls have a positive and rugged feel about them and tend to be manipulated with confidence even by an inexperienced user encountering the product for the first time. We played a variety of discs through the phono section of the C , using our recently acquired Shure V Type IV cartridge for that purpose. Our record library now contains a good selection of direct-to-disc records which were reproduced flawlessly through the C 27 and a high quality 0-watt per channel power amplifier plus our reference loudspeakers. Transient response was superb and bass passages were tight and totally unmuddied. Attempting to use the phone j a c k we discovered that it must be powered from the interconnections of the associated power amplifier. In other words, the C , unlike the C , does not contain its own built-in headphone amplifier but rather derives phone power from the amp-speaker connections which are brought over to the unit for the switching capability incorporated in the C 2 7. A user of the C 27 would want to utilize this feature in any case, and there is an "off" position on the C ' s speaker switch for headphone-only listening once these connections are madeAll in all, Mclntosh seems to have come up with a brand new preamplifier which should fit nicely into those systems which are comprised of the finest power amplifiers, speakers and program source components and whose owners either prefer to do without such ext r a s as graphic equalizers or expanders or who would rather add such devices as separate accessories when and if the need arises. In our opinion, the suggested price of the new C 27 is not at all inconsistent with the level of performance achieved by the C 27 or with its excellent design, construction and control features.

An innovative new lock circuit was developed for use in the MR 80. This new circuit allows correct tuning without the use of a center tune meter. The MR 80 will be correctly tuned regardless of the stations transmitted signal as related to its assigned frequency. Two operational amplifiers cause the lock circuit to track a station even if it drifts 1 MHZ. The mixer is a balanced matched dual J-FET and bipolar transistor circuit. After the mixer the signal is electronically switched to go either directly to the IF amplifier or to go through a quartz crystal filter. The MR 80 has the narrowest IF bandwidth ever used in a stereo tuner. It is the correct width to let just one FM station through. The excellent selectivity of the MR 80 (210 kHz wide at 60 dB down) permits tuning stations that are impossible to receive on ordinary tuners. The SUPER NARROW selectivity position adds a 4 pole - 4 zero crystal filter to the other 5 IF filters. SUPER NARROW permits listening to stations not heard with most other FM tuners. The 5 stages of IF amplification provide the necessary gain to reduce noise in the signal and the interference from other stations. They are piezoelectric fixed frequency filters in place of normal tuned circuits. An advantage the IF stages will always stay in alignment, even with age. The signal strength indicator column is the sum of all IF stages instead of just one it indicates the signal strength throughout the entire IF system. The selectivity section of the IF amplifier is the LIMITER with a total gain of 80 dB. The use of very high gain in the limiter circuit produces hard limiting with very good impulse noise rejection. Limiter bandwidth is greater than 50 MHz, producing excellent detector capture characteristics. A broadband Foster-Seeley discriminator is used
as the detector coupled with a broadband limiter to produce unmeasurable noise and distortion. The heart of the multiplex section is a new third generation phase locked loop (PLL) stereo decoder integrated circuit (IC). This PLL IC incorporates two special systems, an automatic variable separation control circuit to reduce background noise when receiving weak stereo stations, and tri-level digital waveform generation which eliminates interference from SCA signals and from the sidebands of adjacent channel FM signals. The variable separation control is operated from the IF amplifier's signal strength detector. A smooth transition is provided from mono to stereo or from stereo to mono at weak signal levels to provide the optimum signal to noise ratio and best stereo separation for the prevailing signal conditions. The circuit operates only during stereo reception. It switches automatically to monophonic if the 19 kHz pilot tone is absent. Additional advantages of the phase locked loop stereo demodulation are the elimination of inductors to minimize drift, integral lamp driving capability to indicate the presence of the 19kHz pilot carrier, excellent channel separation over the entire audio frequency range, extremely low distortion, low output impedance, and transient-free mono/stereo switching. Following the MPX decoder is the three position de-emphasis switch. The three different positions allow the MR 80 to be used in North America with standard 75ms de-emphasis and in Western Europe and the Far East with 50ms de-emphasis. A 25ms position is provided for use with an external noise reduction devices. An electronic blend filter circuit, using two J-FETs of a quad J-FET package, reduces out of phase noise when in the stereo mode and when tuned to a weak station. The filter is a twin-T bandpass that blends the high and low frequencies, leaving separation unaffected at mid-frequencies. This results in a greatly improved stereo image when the filter is required. Special design attention has been given to the power supply section. Nine separate power circuits are used. Six of these are regulated to prevent loss of performance during a brown out. PERFORMANCE LIMITS and RATINGS TUNING: 88 to 108 MHz ANTENNA INPUTS: One 300W balanced or two 75W unbalanced. INTERMEDIATE FREQUENCY: 10.7 MHz SENSITIVITY: 9.3 dBf (1.6mV) for 35 dB of quieting 14.7 dBf (3mV) for 50 dB of quieting 13.2 dBf (2.5mV) for 3% total noise and harmonic distortion SIGNAL TO NOISE RATIO: 75 dB below 100% modulation

Fig. 3 Mono and stereo quieting and distortion characteristics, FM section.
with input signals of 3.5 uV (dBf) in mono and 30 dBf ( 1 7. 4 uV) in stereo, the stereo result being about the lowest we have ever measured for any stereo FM tuner. At 65 dBf of input signal strength, signal-tonoise ration measured 76 dB in mono (as opposed to 75 dB claimed by Mclntosh) and 71 dB in stereo. Distortion at that strong signal level was the same in mono and stereo, a low 0.1 2% for 1 kHz signal. Distortion (harmonic) versus modulation frequencies, for both mono and stereo operation of the tuner, were with selectivity set to the normal or "narrow" position. We measured specific values of 0.12% at 100 Hz for mono, and 0.1 5% for that frequency in stereo. At 3 kHz, mono distortion was a bit higher than that measured in stereo; 0.18% as against 0.1 3% but at the highest required test frequency of 6 kHz, stereo THD came close to the 0.2% limit while mono distortion was again a low 0.1 %. We should point out that when the "super narrow" selectivity setting is used, distortion does rise rather significantly, approaching the 1.0% mark, but this is a tradeoff that was most deliberately sought by Mclntosh Laboratory's designers. There is just no other way to achieve adjacent channel selectivity of 60 dB without increasing distortion in the stereo mode where sidebands of high frequency modulating signals extend well beyond the single channel width of 100 kHz to either side of center carrier frequency. What is remarkable, in fact, is that Mclntosh was able to achieve this sort of adjacent channel selectivity and still keep the distortion level in stereo under 1.0%! It should be noted, incidentally, that noise performance, or signal-to-noise ratios remain essentially the same in the "super narrow" setting as they were in the narrow or normal selectivity mode.
Fig. 4 Frequency response and separation, McIntosh MR 80 w i t h selectivity switch in "Narrow" (normal position).
Fig 5 - Frequency response and separation, McIntosh MR 8O. with selectivity switch in "Super Narrow" position.

The volume control is a step attenuator which has tracking accuracy within 1 dB throughout its entire range. Such extremely accurate matching is achieved through electronically controlled trimming of the resistance material deposited on pairs of miniature printed circuits. Tracking accuracy and quiet performance are permanently maintained. Use does not affect performance as in ordinary volume controls.
A new third generation phase locked loop stereo decoder incorporates a unique variable stereo
separationnoise reduction control circuit. When listening to weak or distant FM stations, optimum stereo separation and signal to noise ratio is automatically provided. UNIQUE AM TUNER SECTION The AM circuit in the MX 117 is unique in design for a superheterodyne tuner. The AM RF amplifier circuit has constant selectivity, constant sensitivity and high image rejection across the complete AM band. In addition there is no loss of audio frequency response at the low end of the band. It is the only AM circuit that has all these features simultaneously. These advanced AM performance characteristics bring about a re-discovery of AM listening. PERFORMANCE LIMITS and RATINGS PREAMPLIFIER SECTION FREQUENCY RESPONSE; + 0 - 0.5 db from 20 Hz to 20 kHz RATED OUT PUT LEVELS: Main Out 2.5V Line Out 1.25V Headphone 750mV Tape Out 250mV DISTORTION: 0.02% maximum at 2.5V output from 20 Hz to 20 kHz INPUT SENSITIVITY AND GAIN: Input to Main Out Phone 1 and 2 2.2mV in for 2.5V out (61.1 dB gain at 1 kHz) Aux, Tape 1 and 2 250mV in for 2.5V out (20 dB gain at 1 kHz) Input to Line Out Phono 1 and 2 2.2mV in for 1.25V out (55 dB gain at 1 kHz) Aux, Tape 1 and 2 250mV in for 1.25V out (14 dB gain at 1 kHz) Input to Headphone Out Phono 1 and 2 2.2mV in for 750mV out (50.7 dB gain at 1 kHz) Aux, Tape 1 and 2 250mV in for 750mV out (9.5 dB gain at 1 kHz) Input to Tape Out Phono 1 and 2 2.2mV in for 250mV out (41.1 dB gain at 1 kHz) Aux, Tape 1 and 2 250mV in for 250mV out (0 dB gain at 1 kHz) SIGNAL TO NOISE: Phono 1 and 2 -90 dB IHF A-weighted, below 10mV input -80 dB unweighted, below 10mV input Aux, Tape 1 and 2 - 100 dB IHF A-weighted, below 250mV input -90 dB unweighted, below 250mV input INPUT IMPEDANCE: Phono 1 and 2 47kW and 50pF Aux, Tape 1 and 2 47kW OUTPUT IMPEDANCE: Main Out less than 100S7 (to operate into 5kW or greater)
600(2 (to operate into a 600W line) Headphone 8W Tape Out less than 200W (to operate into 5kW or greater) EQUALIZER CONTROL RESPONSE: Center Frequencies 30, 750, and 10 kHz Boost and Cut 12 dB AM SECTION SENSITIVITY: 75mV IHF with external antenna SIGNAL TO NOISE RATIO: 45 dB minimum IHF or 55 dB at 100% modulation FREQUENCY RESPONSE: + 0 -6 dB from 20 Hz to 3500 Hz HARMONIC DISTORTION: 0.8% maximum at 30% modulation ADJACENT CHANNEL SENSITIVITY: 30 dB minimum IHF IMAGE REJECTION: 65 dB minimum from 540 kHz to 1600 kHz FM SECTION USABLE SENSITIVITY: 2mV (11.2dBf) IHF QUIETING SENSITIVITY: 5mV (19.1dBf) IHF -50 dB (Mono) 50mV (39.5dBf) IHF -50 dB (Stereo) SIGNAL TO NOISE RATIO: 70 dB IHF minimum both Mono and Stereo HARMONIC DISTORTION: 0.18% (Mono) IHF maximum 0.38% (Stereo) IHF maximum ALTERNATE CHANNEL SELECTIVITY: 75 dB IHF minimum IMAGE REJECTION: 100 dB IHF minimum STEREO SEPARATION: 45 dB minimum at 1 kHz GENERAL INFORMATION POWER REQUIREMENT: 120 Volts 50/60 Hz, 45 Watts SEMI CONDUCTOR COMPLEMENT: 24 Bipolar Transistors

18 Field Effect Transistors

Line Out

62 Diodes 24 Integrated Circuits MECHANICAL INFORMATION SIZE: Front panel measures 16 inches wide (40.6 cm) by 5 7/16 inches high (13.8 cm). Chassis measures 14 3/4 inches wide (37.5 cm) by 4 13/16 inches high (12.2 cm) by 13 inches deep (33 cm), including connectors. Knob clearance required is 1 1/4 inches (3.2 cm) in front of mounting panel. FINISH: Front panel is anodized gold and black with special gold/teal nomenclature illumination. Chassis is black. MOUNTING: Mclntosh developed professional PANLOC. WEIGHT: 24 pounds (10.9 kg) net, 36 pounds (16.3 kg) in shipping carton.

PRODUCT

ANALYSIS
THE MCINTOSH MX 117 AM/FM TUNER-PREAMPLIFIER
Mclntosh Laboratory Inc. remains one of the few companies that continues to offer a high fidelity component which incorporates both FM/AM tuner circuitry and preamplifier-control circuits. It strikes us that for those audio enthusiasts who do want to include radio (and particularly high-fidelity stereo FM radio) as a program source in their audio systems, this combination, on a single chassis, makes a great deal of sense; so much so, in fact, that we have always wondered why other manufacturers have not offered this type of component in greater numbers. After all, it permits the user to choose any power amplifier he or she requires, allows for the incorporation of two turntables as well as two tape decks, offers the versatility of a separate preamp-control unit while at the same time incorporating a top-performing stereo FM/AM tuner. Front Panel Layout The MX 7 has the unmistakable Mclntosh front panel look, with its anodized gold and black finish, its gold/teal illuminated nomenclature and its exclusive PANLOC mounting system which has become a tradition of Mac equipment. A large, well-illuminated cut-out area of the panel contains the calibrated AM and FM dial scales, and evenly spaced "logging scale" (calibrated linearly in 0.1 increments from 0 to 10), seven small indicator lights (which denote program source selected as well as selection of interstation muting during FM listening) and a fourteen-LED solid-state tuning indicator system, consisting of ten small dots arranged in a vertical row, a small bar which illuminates to denote stereo signal reception and three bars which tell you whether you are tuned above, below, or precisely "on center" frequency when listening to FM. A conventional, flywheel-coupled tuning knob is positioned to the right of the dial area. All remaining front panel controls are neatly arranged along the lower section of the panel. These include a six-position, program source selector switch at the lower left ( A U X , AM, FM, MUTE { F M } , P H { o n o } 1 and PH{ono} 2), three tone controls to the right of center (Mclntosh prefers to call them equalizer controls, since they do divide the audio spectrum into three parts, as opposed to simple bass and treble tone controls which usually control wider swaths of frequencies), a dual-concentric balance and loudness control and a master volume control at the lower right. Six push-button switches plus a stereo headphone jack are located to the left of center of the panel. A light-colored button serves as the power on/off switch while the contrasting dark-shaded buttons handle such functions as TAPE 1 or TAPE 2 monitoring, Tape Copying from either connected tape deck to another, and stereo/mono selection. The loudness control arrangement found on the MX 7 is different from any that we have previously encountered and we shall have more to say about its action presently.

Fig. 1 Front panel view of Mclntosh MX 7 Tuner-Preamplifier MANUFACTURER'S PUBLISHED SPECIFICATIONS: FM Tuner Section: Usable Sensitivity: 1 1.2 dBf ( 2. 0 uV/300 ohms) 50 dB Quieting Sensitivity: 19.1 dBf, mono; 39.5 dBf, stereo. Signal-to-Noise Ratio: 70 dB minimum, both mono and stereo. Harmonic Distortion: 0.18% mono, 0.38% stereo, maximum. Alternate Channel Selectivity: 78 dB minimum. Image Rejection: 100 dB minimum. Stereo Separation: 45 dB minimum at 1 kHz. AM Tuner Section: Sensitivity: 75 uV ( e x t e r n a l antenna) Signal-to-Noise Ratio: 45 dB min IHF or 55 dB @ 100% modulation. Adjacent Channel Sensitivity: 30 dB Minimum. Image Rejection: 65 dB minimum from 540 kHz to 1600 kHz. Frequency Response: +0, 6 dB, from 20 Hz to 3500 Hz. Preamplifier Section: Frequency Response: +0, -0.5 dB, 20 Hz to 20 kHz. Rated Outputs: Main: 2.5 V; Line: 1. V; Headphone: 750 mV; Tape: 250 mV. Distortion: 0.02% max. @ 2.5 V output from 20 Hz to 20 kHz. Input Sensitivity for 2.5 V at Main Out: (Phono) 2.2 mV; (High Level) 250 mV. Signal-to-Noise: (Phono, re: 10 mV in) 90 dB "A" wt'd, 80 dB unwt'd. (High Level, re: 250 mV in): 100 dB wt'd; 90 dB unwt'd. Input Impedance: (Phono 1 & 2): 47 K ohm & 50 pF. High Level: 47 K ohm. Output Impedance: (Main): Less than 100 ohms; (Line): 600 ohms; (Phone): 8 ohms; (Tape): Less than 200 ohms. Equalizer Control Response: dB at 30 Hz, 750 Hz and 10 kHz. GENERAL SPECIFICATIONS: Dimensions: 16" wide x 5 - 7 / " high (panel). Chassis: 13" deep. Knob clearance in front of panel: 1 ". Net Weight: 24 Ibs.

Rear Panel Controls

Stereo pairs of input jacks and a chassis ground terminal are located at the extreme right of the rear panel of the MX 7. Inboard of these are two pairs of "Main" output jacks, line out jacks (designed to operate into 600-ohm loads), the tape-out jacks (for connection to up to two tape decks) and vertical and horizontal oscilloscope output jacks which may be connected to Mclntosh's "Maximum Performance Indicator" or to any oscilloscope for observation of multipath phenomena during orientation and optimization of an FM antenna. A pivotable AM loopstick ferrite antenna is located at mid-panel, and next to it is an FM preselector switch which introduces additional tuned-circuit filtering for additional RF selectivity in the event of strong-signal overload conditions. Normally, this switch is left in the "out" position.
circuit call Automatic Frequency Lock (AFL) which turns on a "lock" voltage when perfectly centered tuning has been reached. The FM I-F section uses five integrated circuit amplifiers and four piezo-electric filters, for a total gain of dB. A full Foster-Seeley discriminator (as opposed to the more common Ratio Detector) discriminator completes the I-F system. The composite demodulated signal feeds the stereo FM multiplex section, the heart of which is a new type of phaselocked-loop stereo decoder IC. This IC incorporates an automatic variable separation control (to reduce background noise during weak-signal stereo reception), and tri-level digital waveform generation which eliminates interference from SCA signals and from the sideband of adjacent channel FM signals. kHz and 38 kHz carrier suppression circuits are used to attenuate any residual carrier components following multiplex decoding. The FM muting circuit employed in the MX 117 operates by detecting ultrasonic noise and by sensing correct center-tuning of the detector circuit. Muting of the audio signal is done with a positive acting FET switching circuit. The AM tuner section employs a three-section tuning capacitor and a special AM-RF amplifier which maintains constant selectivity, constant sensitivity and high image rejection across the entire AM band. An autodyne circuit is used for the AM mixer and two double-tuned transformers are used in the AMIF section. A 10 kHz "whistle filter" is included in the AM tuner section, as is a two-section AVC filter for lower distortion at bass frequencies. The phono preamplifier-equalizer section uses an IC operational amplifier whose differential input stage has been optimized for low noise and low distortion. The feedback network which also provides RIAA equalization employs 1 % metal film resistors and 5% polyfilm capacitors. The gain of this preamp section is just over 41 dB. As for the unusual loudness control arrangement referred to earlier, it uses the same sort of IC operational amp used in the phono preamp stages. Two feedback loops are employed: one flat, the other conforming to the Fletcher-Munson equal loudness contours. A potentiometer (the front panel "LOUDNESS" control) placed between these loops makes it possible to select any curve from flat response to full loudness compensation. Once the contour is set by the user, it remains fixed and independent of the position of the master volume control. The equalizer-amplifier also uses a low-noise operational amplifier. Three other op-amps are arranged in the circuit equivalents of three tuned circuits; each resonant at one of the three equalizer center frequencies. The output amplifier of the MX 7 is a push-pull complementary class AB circuit which uses a signalinverting differential stage at its input. The amplifier drives the main and line outputs as well as the head-

Fig. 6 Frequency response (upper trace) and stereo separation, Mclntosh MX 7 Tuner-Preamp Figure 6 is a 'scope photo of a spectrum analyzer multiple sweep in which the upper trace represents output from the left channel main output with a Leftonly signal modulating our FM generator and sweeping from 20 Hz to 20 kHz. The lower trace was obtained by subsequently measuring the output of the Right channel under the same modulation conditions, and is therefore a measure of stereo separation versus frequency. The vertical scale is dB per
division in this and all other 'scope photos. At 1 kHz, separation measured an impressively high 55 dB fas opposed to Mclntosh's conservatively guaranteed 45 dB) while at the frequency extremes of 100 Hz and 10 kHz we measured 37 dB and 35 dB respectively. In Fig. 7, the sweep mode was changed, so
the listener to enjoy reception of marginally quiet signals while still benefitting from the interstation muting feature. We meaured an alternate channel selectivity of 80 dB for this sample. SCA rejection, under conditions of a modulated 67 kHz SCA subcarrier injected at 10% of total modulation of the main carrier was a very satisfactory 60 dB below reference output level while stereo sub-carrier product rejection was in excess of 65 dB. While it is not our practice to spend too much time testing the AM sections of AM-FM tuners (most of them are simply so poor in performance that they are not worth bothering about), we do, as a matter of course, measure at least the frequency response of the AM tuner sections. As can be seen in Fig. 8, we were pleasantly surprised to find absolutely flat response down to 20 Hz (most AM tuners tend to roll off bass severely below 50 to 100 Hz or so). And while response to 3.5 kHz may not seem like "hi-fi" reproduction to most listeners, it is actually better than the response obtained from most of the AM sections of combination AM-FM tuners or receivers that we measure in our laboratory.
Fig. 7 Crosstalk components, 5 kHz, Mclntosh MX 117 Tuner-Preamp that this time the sweep was linear, from 0 Hz to 50 kHz, with the lightly visible scale corresponding to 5 kHz per horizontal division. The tall spike at left is a 5 kHz output from the modulated channel. Contained within that spike is the representation of the opposite (unmodulated) channel output, while to the right of these are various cross-talk products at harmonics of 5 kHz as well as any residual kHz or 38 kHz sub-carrier output products. All of these extraneous output products were at least 60 dB or better below the desired reference output level. Note, too, that separation at the relatively high 5 kHz frequency was approximately 40 dB (the difference in "height" between the two spikes at 5 kHz-one within the other-at the left of the display). Muting threshold for this FM tuner was set to dBf (3.9 uV), an ideal signal level for this type of muting circuit, since it allows

Fig. 9 Range of equalizer controls, Mclntosh MX 117 Tuner-Preamp Preamplifier and Control Section Measurements Mclntosh Laboratory has chosen to publish their specifications relating to the audio portions of the MX 7 in a way that pre-dates the new IHF/EIA Amplifier Standards, whereas in our lab we adhere to the newer standards. Since this would make it difficult to compare published specs with measured results, we decided to make both types of measurements. In that way, those interested in comparing results with those obtained for other products where the IHF/EIA standard was used will be able to do so, while those wishing to compare results with Mclntosh published specs can do so as well. IHF/EIA phono input sensitivity measured 0.45 mV. This corresponds to almost exactly 2.2 mV as claimed by Mclntosh for the higher 2.5 volt output. As for the high level inputs, 50 mV of signal was required to deliver the
Fig. 8 Frequency response, AM section, Mclntosh MX 117 Tuner-Preamp
reference 0.5 volts output, and this too corresponds exactly to the 250 mV spec called for by Mclntosh for their referenced output level of 2.5 volts. Figure 9 illustrates the range of control of each of the three equalizer controls provided on the MX 117. The frequencies at which the three equalizer controls are centered make for an extremely versatile range of control. The action of the unique loudness control found in the MX 7 is illustrated in the series of response curves plotted by means of our spectrum analyzer and reproduced in the 'scope photo of Fig. 1 0. Note that only a moderate amount of treble compensation comes into play as the control is advanced towards more contouring.
Fig. 12 Expanded-scale phono response, using inverse RIAA input test signal from 20 Hz at the left to 40 kHz, with double vertical lines indicating 1 00, and 1 0,000 Hz. 0 dB reference has been established at 1 kHz as shown. In Fig. 12 the vertical sensitivity of the plot has been expanded so that it is 2 dB per vertical division (instead of 10, as in the previous display). Also, the signal applied to the phono inputs has not been subjected to inverse RIAA equalization so that, in theory, if perfect RIAA equalization were incorporated in the MX 7, a "straight line" response would be obtained. As you can see, maximum deviation from this ideal was 0.6 dB at 40 Hz (where the dotted line cursor is positioned for a read-out at the lower right of the display) while moving the cursor to the maximum deviation in the high frequency region (Fig. 13), shows a maximum deviation from "absolute RIAA accuracy" of only 0.5 dB at 1 8.5 kHz.