Reviews & Opinions
Independent and trusted. Read before buy NAD L53!

NAD L53 Manual

Preview of first few manual pages (at low quality). Check before download. Click to enlarge.
Manual - 1 page  Manual - 2 page  Manual - 3 page 

Download (English)
NAD L53 Dvd Player, size: 1.5 MB
Instruction: After click Download and complete offer, you will get access to list of direct links to websites where you can download this manual.


NAD L53About NAD L53
Here you can find all about NAD L53 like review and other informations. For example: receiver, manual.

NAD L53 manual (user guide) is ready to download for free.

On the bottom of page users can write a review. If you own a NAD L53 please write about it to help other people.
[ Report abuse or wrong photo | Share your NAD L53 photo ]

User reviews and opinions

No opinions have been provided. Be the first and add a new opinion/review.




The Audio Technology Authority
Article prepared for
NAD L53 DVD/CD Stereo Receiver

By Charles Hansen

PHOTO 1: Front view of unit with remote control.
he L53 is a 50W per channel stereo AM/FM receiver that also includes a CD/DVD player. The only other gear required are stereo speakers and a TV monitor. You can dispense with the monitor if you only want to listen to audio. The L53 includes a learning remote with illuminated keys that you can program to replace your TV, and/or cable and satellite remotes, and you can even program it to activate your entire system with only two keystrokes. A clock/timer function is included in the front panel display. The L53 uses the latest generation DVD-video decoder with progressive scan output to a digital TV display through its component video connections. Composite and S-video connectors are provided as well, although an RF (modulator) output is not available. If your TV is an older model without A/V input jacks, you can always use a VCR as the RF modulator. The CD/DVD drive will play DVD-V, VCD, SVCD, CD, CD-R, CD-RW, MP-3, JPEG, and WMA data. The L53 will not play DVD-Audio or SACD high-resolution audio formats. The L53 features Sound Retrieval System (SRS) technology to provide a surround sound simulation from only two speakers in rooms that cannot accommodate additional surround speakers. A subwoofer output that tracks the stereo level control allows full compat-
ibility with any of the many powered subwoofers currently available. The stereo power amplifier section is a discrete transistor design, and all the power supplies use conventional iron transformers and linear regulators rather than the more common computer-style switching power supplies. As such, the L53 is a hefty 18.5 lb.


Photo 1 shows the front panel of the L53 along with its HTR-L53 remote control. The power switch is located on the left side just below the blue LED standby indicator. Pressing the drive open/close button will also turn the unit on. After a startup delay, the standby indicator will go out and the main display will come on. There is no off position, so when the power switch is cycled off the unit remains in standby mode. The remote control infrared sensor is to the right of the power switch. The access door just below the power switch covers a " headphone jack and the Video 4 composite and S-video input jacks. When headphones are plugged in, the speaker outputs are muted.
To the left of the drive tray are four slanted buttons that control the clock/ timer, the FM mute/mode button, the setup/memory button (used to store up to 60 station presets or access the DVD options), and the display format button. The blue vacuum fluorescent main display is located just below the drive tray. To the right of the drive tray are the play/pause button, the drive stop/open button, the audio mode select, and the input select. The four buttons below them control the forward/back/skip/ scan/tune functions. The large knob on the right is the volume control, which is not a dual analog pot, but a servo control that also actuates the bass, treble, and balance controls in conjunction with the setup menus. The HTR-L53 is an eight-device illuminated learning remote control with macro function. The manual is a bit thin with regard to all its features, but it seems straightforward enough, and the lighted color-coded buttons are a welcome respite from the sea of blackbuttons-on-black-background-with-tinyprinting all too common in remotes these days.

audioXpress 2005 1

nectors appear to be nickel- plated. To the right of all these connectors are a Xantech-style infrared repeater input, S/PDIF coax and optical PCM digital inputs, and an PHOTO 2: Rear view of unit. optical PCM outThe L53 is constructed of heavy put jack. Just below these connectors gauge gray painted steel, with cool- are SCART/RGB connectors for the Euing slots on the top and bottom. The ropean PAL television signals. Finally, front plate is 10mm-thick aluminum the three vertical connectors on the rather than the plastic used on many of right side of the connector stack supthe consumer A/V products. Four thick ply component video Y/Cb/Cr signals to hemispherical rubber mounting feet video monitors that accept them. provide ample finger room for lifting The polarized two-prong power cord the L53. is permanently attached, with a polarFacing the rear panel (Photo 2), the ized two-prong accessory outlet rated left side has two pairs of speaker bind- for 100W/1A just below. ing posts. In accordance with CE rePhoto 3 shows the amplifier with the quirements, these are not on " USA cover removed. A schematic was not spacings, so you cannot use dual ba- furnished with the unit, but service nana plugs. Each threaded post has a manuals are usually available from chassis-mounted plastic guide bush- NAD at extra cost. Looking just inside ing that accepts heavy gauge wire. The the left rear panel, youll note the power manual says speakers should be rated cord is threaded through a toroidal ferat 8 minimum. rite EMI filter core. The left rear of the Next in line are the clip terminals L53 has an AC power input board with for the supplied AM loop antenna and a MOV transient suppressor, film caps, a 75 F-connector for the FM input. A and the line fuse. 300 dipole is supplied along with a The small transformer supplies matching 300: 75 balun. power to the front panel display/control Composite and S-video input connec- board just under the front lip of the tors are supplied for a cable/satellite chassis (out of view in the photo). The box and a VCR. There are also compos- larger transformer at the left front supite and S-video output jacks that you plies the high current requirements for can use to send A/V signals to a VCR the power amplifier, and the lower voltor DVD recorder. Next to the A/V jacks ages needed for the logic, video, digital are the subwoofer output jack and the audio, and analog line-level audio circomposite and S-video monitor output cuits. The location of these two power connectors. All the RCA jacks have transformers makes the L53 much gold-plated shells, but the center con- heavier on the left side. F ront and cenPHOTO 3: Interior view. ter is a Raymedia CD/DVD drive unit with an A/V control/inter face PC board incorporated in the bottom of its housing. The DVD format is handled by an ESS Vibratto 2 MPEG-4 chip set with a 54MHz 10-bit video DAC. A Cirrus Logic CS4391A 24-

bit/96kHz stereo DAC decodes 16-bit PCM to analog audio. Stacked just behind the drive are two horizontal PC boards. The bottom board is the video interface and switching board, while the audio and logic control is on top. The various functions are controlled by means of an STMicroelectronics ST92F124V1T6 8/16-bit microcontroller chip with 128K of flash memory. A minimal number of discrete 74ACT chips provide TTL-to-CMOS glue logic. The analog audio is processed through an NJR NJW1153P six-channel electronic volume control IC with input select, tone control, mute functions, and three output record channels. Based on the data sheet for this chip, it appears to be entirely analog in the signal path. A Philips SAAA6579T chip demodulates the Radio Data System (RDS) information from the tuner section. An SRS Labs SRS31D0 chip functions as the Sound Retrieval System (SRS) surround ambient decoder. There are two dual op amps on the audio board. One is an NJM2068M, roughly identical to ye olde RC4558 from the mid-1970s. The other, despite the fact that the silk-screen ID in its location also reads 2068M, is a highquality Burr-Brown OPA2134 located adjacent to the left/right channel audio inputs to the power amplifier board. The 2068 op amp appears to provide summing and low-pass equalization for the subwoofer output. Dual supply voltages are provided for the NJW1153 by L7808 and L7908 linear regulator ICs. The rear panel audio input and output jacks are all coupled with aluminum electrolytics. The video board uses discrete RF transistors for video output buffering. Additional regulated DC supplies (+5, 12V) for the audio board come from the front area of the power amp supply board on the right side of the unit. The three linear regulator ICs are located on the forward edge of the heatsink. The power amplifier DC supply is at the rear of the supply board. Main reservoir capacitors are two 6800F 50V Koshin electrolytics. Supply rail fuses and the speaker Zobel networks and output inductors are located nearby. Just in front of the speaker binding posts is a 5A startup delay/

2 audioXpress 2005
protection relay. The discrete power amplifier circuitry is on the vertical PC board attached by metal clips to the large aluminum heatsink. The AM/FM tuner and PLL are located in the shielded enclosure directly in front of the antenna jacks. All-in-all, the L53 has a compact, well-thoughtout component arrangement. NAD also makes the L73 Surround Sound Receiver with five channels of audio amplification and adds DVD-Audio playback and HDTV-compatible video output.
One nice feature in both CD and MP3 mode is the ability to go directly to track 1 while the last track on a disc is playing by pressing the next track button. Not all CD players will do this, and require you to enter track 1 on the numeric keypad.

Because the L53 includes AM/FM tuners and the DVD/CD/MP3 player in one package, it is unlikely that owners would use the composite audio input jacks for other stereo sources, such as a cassette deck or outboard phono preamp. However, all the internal and external source options need to flow through the NJW1153P six-channel control IC, so its performance will set the baseline for the audio performance of the L53. There is very little in the way of important audio parameters in the NJW1153 data sheet, other than gain (always near unity) and THD+N curves versus input voltage. Important specifications such as slew rate, unity gain bandwidth, input impedance, bias current/voltage, and noise performance go missing. I put the L53 in CD playback with a CD test disc for 1 hour before making any line level measurements. The player was just warm to the touch above the heatsink after this period at 1W into 8. Input impedance for the CAB/SAT and VCR audio line inputs was approximately 90k. The S/PDIF digital coax input impedances measured the specified 75. Output hum and noise with the analog inputs terminated in 600 measured 89dB (ref 2V RMS), decreasing to 107dB A-weighted. Front left and right channel separation measured a fine 80dB at 10kHz, unweighted. The output impedance for the twochannel stereo VCR outputs measured 460 from 1kHz to 20kHz, and a higher 700 at 20Hz where the output coupling capacitor adds to the impedance. L-R channel balance was essentially perfect. The stereo outputs do not respond to variations in the volume, balance, or tone control settings, and are passed through from the sources at nearly unity gain (0.26dB), although this will change with load. The subwoofer output does follow
the volume control setting, and had an output impedance of 110, and showed unity gain with a volume control indication of 1 (100k load). The S/PDIF digital coax and the three component video output impedances are all specified to be 75. All input and output jack shells and the speaker negative binding posts are connected to chassis ground.

Frequency response for line level sources (CD, MP3, FM tuner, and the VCR line input) into 100k is shown in Fig. 1. The VCR line level output is also shown with a 10k load (dashed line). The 1kHz 0dBfs CD line output was 2.12V RMS at 1kHz, or 0.60dB higher than the CD Red Book standard of 2V RMS. There was just 0.2dB of de-emphasis response error at 1kHz when playing pre-emphasized CD test tracks. The MP3 frequency response with identical CD test tracks ripped to 128k MP3 format was 0.38dB higher through the midband before dropping off just above 15kHz. The subwoofer output response is shown with 2V RMS input and the volume at 2 to improve clarity on the graph. I measured the FM tuner frequency response with an RF signal level of 65dBf and 100% (15kHz) modulation. In this measurement there are three responses that can contribute to the overall error: the 75S pre-emphasis and steep 16kHz LP filter on the audio that is fed to the FM signal generator, and the 75S de-emphasis in the tuner under test. FM stereo crosstalk performance was 42dB at 1kHz. Note that the FM stereo line level output at 1kHz with 100% modulation is 750mV, or 8dB relative to 2V RMS. Accordingly, the FM frequency response curve has been shifted up by +6dB to keep it within the graph.


Before making any measurements, I set up the L53 for some selected listening to make sure all the audio modes were operating properly. It was at this point I found that the included remote control batteries had not been included, and that the L53 was very easy to operate from its front panel controls. When you insert a disk, the unit must determine the format before playback begins, because all shiny silver 120mm discs initially look the same. It took about 15 seconds from the LOADING display until the format was recognized and accepted for play. Initially, the drive (Korean-sourced, but made in China like the rest of the L53) was fairly noisy up close to the unit, but the noise gradually subsided until it was no louder than my Rotel CD player. Because the L53 can also play MP3 files, I listened to the 128K-sampled compressed MP3 CD-R that I had previously made for the Pioneer DV-563A test (aX November 2004, p. 36). My perception was that the MP3 tracks were a bit louder than the CD tracks of the same songs. Imaging was better than with the DV-563A, but the soundstage was still flat, with most of the reverb or ambience that was present in the CD version lost in the MP3 encoding. This actually improved the sound in some pop music songs where various effects were applied to excess in the original CD tracks. Initially, even though the total time and track count were correct, I couldnt access all the MP3 tracks. Once track 1 began playing, I was able to access any track skip/scan tracks. I dont know whether this was peculiar to MP3 or due to the CD-R media. I didnt see this behavior with commercially pressed CDs.

FIGURE 1: Frequency response (line level sources).


audioXpress 2005 3
THD+N versus frequency is shown in Fig. 2 for each audio line level mode. I engaged the 22kHz LP filter in my distortion test set for the CD and MP3 tests, and the 80kHz LP filter for the line input and FM tuner. The analog CAB/SAT input voltage was held to 2V RMS because this is the 0dB level on the CD test disc and the VCR line output does not respond to the volume control. Distortion for the FM tuner section is shown for mono and stereo. The analog line input distortion with a 10k load is again shown with a dashed line. The THD+N for the MP3 test tracks shows a bit of a roller-coaster ride. There is also a dip at 60Hz that may in-
dicate that some 60Hz AC line pickup is being notched out by the distortion test set at that frequency. THD+N vs. input voltage is shown in Fig. 3, at 1kHz for each line level audio mode into 100k (I excluded the FM curve because reducing the modulation level greatly increased the distortion, which was near 0.5% to begin with). I measured the subwoofer THD at 50Hz. This time I added a very low 600 line level load at 1kHz (dashed curve).
The residual distortion signal with an analog input of 1kHz at 2V RMS into 10k is shown in Fig. 4. The upper waveform is the VCR line output signal, and the lower waveform is the distortion test set monitor output after the test set notch filter, not to scale. THD+N at this test point is a low 0.0029%. Note the
FIGURE 2: THD+N vs. frequency (line level sources).
FIGURE 5: Line input 1kHz residual distortion with 22Hz LP filter.


FIGURE 3: THD+N vs. input voltage (line level sources).


FIGURE 6: Line input spectrum of 50Hz sine wave.

R-2596-03 R-2596-06

FIGURE 4: Line input 1kHz residual distortion with 80kHz LP filter.
FIGURE 7: Line input extended spectrum of 1kHz sine wave.


series of 44.1kHz spikes riding on the low-level noise, perhaps due to coupling from the digital audio section (although the CD player is off at this time). Figure 5 shows the same data, but with the steep 22kHz low-pass filter rather than the gradual 80kHz low-pass filter engaged. This confirms that the spikes do not produce any noticeable artifacts within the audio band. The negative half-cycles appear to have a predominant second harmonic, while the positive half-cycles are a mix of higher harmonics and noise. The spectrum for a 50Hz sine wave at 2V RMS into 100k is shown from DC to 1.3kHz in Fig. 6. The THD+N measures only 0.0024%. The levels of the second through fifth harmonics were 108dB, 96dB, 98dB, and 103dB, respectively, for a calculated THD of 0.0021%. Decreasing the load to 10k produced a uniform increase in all the harmonics for a THD+N reading of 0.0061%. Repeating the spectrum at 1kHz (not shown) saw a similar distribution of harmonics with a THD+N of 0.0031% into 100k and 0.0039% into 10k. The slight rise in THD near 60Hz in Fig. 2 may be caused by low-level coupling of the power line, but it was not visible in Fig. 6. In view of the presence of coherent HF spikes in the residual distortion of Fig. 4, I extended the 1kHz spectrum to 166kHz to see whether they were present (Fig. 7). Here you can see the 44.1kHz spike at 90dB and an additional 88.2kHz spike at 95dB. Again, there are no visible artifacts in the audio band, but the presence of these two frequencies with the CD player off is puzzling, if benign. The logic clock remains running for the microcontroller, of course. Figure 8 shows the L53s output spectrum reproducing a combined 19kHz + 20kHz CCIF intermodulation distortion (IMD) signal just below the 6.8V clipping level of the NJW1153 chip into 10k. The 1kHz IMD product is 96dB (0.0016%), with the 18kHz and 21kHz products at 85dB, or 0.0056%. Repeating the test with a multi-tone 9kHz + 10.05kHz + 20kHz IMD signal (Fig. 9) resulted in a worst-case 18kHz product of 79dB or 0.011%. There is also a picket fence of low-level 1kHz products throughout the spectrum.

4 audioXpress 2005

I viewed the response of the L53 to three square-wave test frequencies using a 10k load. The response at 40Hz (Fig. 10) showed a moderate and acceptable amount of tilt, not all that different from the same 40Hz into 100k. The 1kHz square wave (not shown) had a bit of rolloff on the positive-going edges but a single half-cycle peak at the negative-going edges. This is more apparent with the 10kHz square wave shown in Fig. 11. Because there is no high-frequency response peaking for the analog input signals into resistive loads in Fig. 1, I cant really explain this asymmetry. I ran it long enough to ensure it wasnt due to aliasing with my DSO.
altered by the tone controls. Matching subwoofer level with the main speakers can be performed using the level control on a powered sub, after which the L53 subwoofer line output level tracks the volume control in the L53. The NJW1153 chip incorporates a subwoofer level control that, like the other five channels, tracks the six-channel Amp Gain setting, but the chip does not have any subwoofer low-pass crossover EQ capability. I surmised earlier that the subwoofer stereo summing and LP filter functions are performed by the 2068 op amp located adjacent to the subwoofer line output jack.


The subwoofer line output frequency response (Fig. 1) is fixed and cannot be


I performed measurements on the L53 CD player section using my usual test discs. I did not attempt to measure the Dolby Digital or DTS audio per-
FIGURE 8: Spectrum of 19kHz + 20kHz CCIF intermodulation signal.
FIGURE 11: Square-wave response at 10kHz into 10k.

R-2596-08 R-2596-11

FIGURE 9: Spectrum of 9kHz + 10.05kHz + 20kHz multi-tone intermodulation signal.
FIGURE 12: Distortion residual of CD 1kHz sine wave.

R-2596-09 R-2596-12

FIGURE 10: Square-wave response at 40Hz into 10k.
FIGURE 13: Spectrum of CD 50Hz sine wave.


formance because I do not have any DVD-V test discs. The analog VCR line outputs showed normal polarity, a positive-going test pulse producing a positive-going output. The digital black test tracks measured 100dB, indicating that the L53 was probably muting the outputs during this test. The channel separation test tracks consist of a 0dB sine wave on one channel, and digital black on the opposite channel whose crosstalk is being measured. The L53 also appears to mute the analog output of the silent channel during this test. I measured the response at the VCR audio line output jacks, as I did with the other sources. In CD mode, the player performed perfectly in the track defect dropout tests out to track 32 on the Pierre Verany test disc, which contains a 1.25mm gap (Red Book requirement is 0.2mm). Track 33, with a 1.5mm gap, produced clicking sounds and some audio muting. The residual distortion signal in CD mode for 1kHz at 0dBfs (Fig. 12) shows the same harmonic asymmetry as the line level residual in Fig. 5. The MP3 residual distortion signal (not shown) was similar, with a bit higher noise level. The spectrum of a CD 50Hz sine wave at 0dBfs is shown in Fig. 13, from DC to 1.3kHz. The calculated THD based solely on harmonics was 0.0078%, while the THD+N measured 0.009% with the distortion test set. It stayed at this level through 1kHz before rising at higher frequencies as shown in Fig. 2. The second harmonic measured 91dB, the third was 96dB, the fourth 115dB, and the fifth at 82dB. Since the CD spectrum is characterized by a much higher noise level than that of the line input alone in Fig. 6, I trust that the spectrum analyzer has properly isolated the actual harmonics hidden within the noise. There are no 60Hz power line harmonics or other spuria evident in the spectrum. The MP3 spectrum in Fig. 14 shows a higher concentration of noise along with harmonically unrelated spikes as high as 73dB. The THD+N measures 0.11%, although it drops to 0.034% at 1kHz (Fig. 2). Figure 15 shows the CD spectrum of response to equal level CCIF 11kHz and 12kHz intermodulation test signals,

audioXpress 2005 5

each at 6dBfs, from DC to 20.8kHz. The 1kHz IMD difference product measures 72dB with a matching product at 2kHz. There are no products evident at 10kHz or 13kHz, but they may be buried in the noise. When the same IMD signals are ripped to MP3 format, the results are not quite as good (Fig. 16). The 1kHz and 2kHz products are higher at 71dB and are joined by visible 10kHz and
13kHz products. The rising skirts around the 11kHz and 12kHz stimulus signals indicate noise that may be due to the MP3 audio compression. There are also additional spikes that have no relation to the IMD test signals. A repeat of the test in CD mode with the more difficult 19kHz + 20kHz IMD test track (Fig. 17) shows the 1kHz, 18kHz, and 21kHz intermodulation difference products to be about 80dB
FIGURE 14: Spectrum of MP3 50Hz sine wave.
FIGURE 18: Undithered CD 1kHz sine wave at -90.31dBfs.


FIGURE 15: Spectrum of CD 11kHz + 12kHz intermodulation.
FIGURE 19: 1/3 octave spectrum of CD 1kHz sine wave at -90.31dBfs.


FIGURE 16: Spectrum of MP3 11kHz + 12kHz intermodulation.
FIGURE 20: 1/3 octave spectrum of MP3 1kHz sine wave at -90.31dBfs.
(0.01%). But for the noise level, this is decent performance. Figure 18 shows the CD reproduction of an undithered 1kHz sine wave at 90.31dBfs. At this level the signal consists of 1 bit of data, producing two different voltage levels that are symmetrical about the horizontal axis (time). These discrete voltage steps are obscured in the 80dB noise floor, so the effective bit rate is probably about 14 bits. Noise also took its toll on the octave spectrum of the same undithered 1kHz sine wave at 90.31dBfs shown in Fig. 19. There is a slight rise in the spectrum around 120Hz, which may be a power supply rectification artifact, but a discrete 1kHz peak is nowhere to be found. A octave spectrum of the 90.31dBfs test track ripped to MP3 showed a series of peaks at multiples of the 60Hz line and higher initial noise floor (Fig. 20). The CD playback of a 0dBfs square wave at 997Hz (Fig. 21) exhibits a slightly unsymmetrical Gibbs phenomenon damped sinusoid ringing associated with the steep digital filters used in the L53. Reproduction of the same square wave ripped to MP3 is shown in Fig. 22. Note that the Gibbs pre and post echo ripple has a total of 16 pulses per full cycle, compared with the 22 pulse CD playback. You can approximate the PCM Nyquist frequency response limit by multiplying the number of pulses by the fundamental square-wave frequency. This demonstrates the more limited high frequency response available as a result of the MP3 compression algorithm. The MP3 waveform is even less symmetrical, and appears to be slightly modulated by 60Hz.


FIGURE 17: Spectrum of CD 19kHz + 20kHz intermodulation.
FIGURE 21: 997Hz CD square-wave response.
I did not run any tests on the AM section of the tuner, other than making sure it was functional. The FM tuner has two modes: FM and MUTE. The display shows Tuned when the tuner locks to a valid FM signal, although there is no meter or bar graph to show signal strength. The L53 FM tuner inverts the audio polarity. The L53 stops its manual station scan every 100kHz, even though only



6 audioXpress 2005

the odd 0.1MHz channels are assigned in North America. Same reason that the SCART video output connector is provided, I guess, except these additional FM frequency stops are a bit more time consuming and less convenient to the user during manual tuning. (Manufacturers Note: the SCART connector is supplied in the European version only). The L53 tuner quieting characteristics are shown in Fig. 23. I needed to switch to FM mode (not MUTE) to measure mono sensitivity. The minimum subjective usable sensitivity using manual tuning was 10.6dBf, with some visible noise breakup in the audio signal on the scope. The mono signal became free from visible noise artifacts at 16dBf. The station auto search threshold, where it would stop at the signal generators frequency during a station scan, was 19dBf. The red stereo indication came on at 22dBf. The 50dB stereo quieting signal strength was approximately 30dBf. There was no overload at the maximum RF input of 100dBf. The audio output disappeared when I moved the FM test signal 200kHz to either side of the tuned center frequency, so adjacent channel rejection should be adequate. When I engaged the FM MUTE mode, the auto search threshold increased to 34dBf. Using the supplied FM dipole antenna and 300 to 75 balun, I could pick up only about half the FM stations in FM MUTE as in FM mode with its lower auto search threshold. Performance was much better in my audio system with its dedicated 75 FM antenna. The FM performance is certainly adequate, especially in a strong signal area, but not quite as good as my venerable NAD 4155 dedicated tuner.

showed a large increase in crossover spikes, so I assume the amplifier bias was decreasing to reduce temperature. I lowered the power level to 1W and after another 15 minutes things returned to normal. Admittedly, a continuous 10W sine wave is a difficult operating condition. It is close to the power or 16.7W that produces maximum dissipation in the output transistors. The measurements for the right channel are presented here. There is no noise at all when starting up or shutting down the amplifier. Output hum and noise measured 89dB, input shorted (109dB with A-weighting), and some hiss was audible at full volume with my ear against the speaker. I also measured 40mV (L) and 13mV (R) of DC offset. The L53 does not invert overall (line in to speaker out) polarity. The gain at 2.83V RMS output into 4 and 8 loads was a high 35dB and 36dB, respectively. I noticed that the right channel output was about 0.5dB higher throughout the entire volume control range, while the line level outputs tracked perfectly. The output impedance at 1kHz was 0.12, increasing only slightly to 0.13 at 20kHz. The frequency response at 2.83V RMS for loads of 4, speaker (dashed line), 8, no-load, and 8 paralleled with 2F (bottom to top at 30kHz) is
shown in Fig. 24. Response was within 1dB from 10Hz to 57kHz, at a reference output of 2.83V RMS at 1kHz into 8. It wasnt down to 3dB until 110kHz. There was just a bit more HF rolloff with a 4 load. The response with a complex load of 8 paralleled with a 2F cap (a test of compatibility with electrostatic speakers) produced a pronounced high frequency peak exceeding +5dB at 70kHz, at which point the L53 tripped into its protection mode. The flat response in the audio band tells me the L53 amplifier will be insensitive to variations in speaker impedance with frequency, although it may protest electrostatics. At low frequencies the noise floor limited the excellent crosstalk performance, but there is adequate isolation between the stereo channels at higher frequencies (Table 1). The tone control frequency response range is shown in Fig. 25. The tone controls only operate on the power amplifier output and are inactive at the line or subwoofer outputs. The maximum spread at the low and high frequency extremes exceed the specified 10dB by a few dB. I made all the amplifier distortion measurements with the tone controls

TABLE 1: Crosstalk.

Frequency 100Hz 1kHz 10kHz 20kHz R to L 77dB 77dB 73dB 71dB L to R 77dB 76dB 72dB 70dB
FIGURE 22: 997Hz MP3 squarewave response.
FIGURE 24: Frequency response.


I operated the L53 amplifier with a line input signal of 0.5V RMS 1kHz and adjusted the volume for 10W into 8. For most of the intended one-hour period, the THD for both channels remained near 0.01%. The temperature above the case cooling slots became quite hot after this period, measuring 58 C. After 45 minutes, the THD suddenly rose to 0.15%. The distortion residual

FIGURE 23: FM quieting - NAD L53.


FIGURE 25: Tone control frequency range.



audioXpress 2005 7

flat and the SRS turned off. In addition to the fixed test loads, I added an NHT SuperOne loudspeaker1 for this measurement. THD+N versus frequency is shown in Fig. 26 for outputs of 8 10W, 8 1W, 8 paralleled with 2F, 4 2W, 8 33W, and 4 10W (bottom to top
FIGURE 26: THD+N vs. frequency.
at 20kHz). The speaker load is shown with a dashed line, and its distortion is inversely proportional to the speaker impedance. The impedance and phase of the NHT SuperOne is shown in Fig. 27. During distortion testing, I engaged the test set 80kHz low-pass filter to limit the out of band noise. Figure 28 shows THD+N versus output power for the loads and frequencies shown in the graph. The L53, with both
FIGURE 30: Residual distortion 1kHz 33W.


FIGURE 27: Speaker impedance and phase.


FIGURE 31: Spectrum of 50Hz line input sine wave.


FIGURE 28: THD+N vs. output


FIGURE 32: Spectrum of 50Hz CD sine wave.


FIGURE 29: Residual distortion 1kHz
FIGURE 33: Spectrum of 19kHz + 20kHz intermodulation signal.


channels driven, reached its 1% clipping point at 55.8W with the 8 load (for 0.47dBW of headroom) and 73.4W with the 4 load (1.67dBW). The positive half-cycles clipped just slightly before the negative half-cycles, and both showed a scoop-out at the peak of the sine wave as if the top 5% of the peak were inverted back into the waveform. This was a bit unusual, as most solidstate amplifiers clip flat across the top. The chassis above the heatsink reached a maximum temperature of 61 C after 15 minutes at 16.7W into 8, so I reduced power to avoid another protective trip. The distortion residual waveform for 5W into 8 at 1kHz is shown in Fig. 29. The upper waveform is the amplifier output signal, and the lower waveform is the monitor output (after the THD test set notch filter), not to scale. This distortion residual signal shows mainly noise with perhaps a hint of higher order harmonics. THD+N at this test point is a low 0.01%. Increasing the output power to 33W (Fig. 30) showed a bit of periodic scoopout in the noise, suggesting second harmonics on the negative half-cycles. Distortion here is even lower at 0.0082%. The spectrum of a 50Hz line input sine wave at 5W into 8 is shown in Fig. 31, from DC to 1.3kHz. The second, third, fourth, and fifth harmonics measure 87dB, 90dB, 105dB, and 93dB, respectively. The calculated THD based on the distribution of harmonics is 0.0053%, while the THD+N measured 0.01%. Repeating the 5W spectrum with the 50Hz 0dBfs CD test track (Fig. 32) produced a spectrum with a higher noise floor. THD+N is also higher at 0.023% with a more prominent second and fifth harmonic. Figure 33 shows the amplifier output spectrum reproducing a combined 19kHz + 20kHz CCIF IMD signal at 20Vpp into 8. The 1kHz IMD product is 82dB (0.008%), and the 18kHz product is 79dB (0.011%) with the 21kHz product just off the graph. Repeating the test with a multi-tone IMD signal (9kHz + 10.05kHz + 20kHz, shown in Fig. 34) resulted in 950Hz and 1050Hz products of 82dB, a group of spikes around the two lower tones and above 15kHz of 95dB, and an 18.95kHz product of 92dB. This test gives a better

8 audioXpress 2005

indication of the amplifiers nonlinear response, because it is a closer approximation to music than a sine wave. The L53 power amplifier produced good IMD results, even better than the line input signals of Figs. 8 and 9, which were taken at a higher voltage and not at the lower level that reflects the additional voltage amplification by the power amplifier. The 2.5Vp-p 40Hz square wave in Fig. 35 shows more tilt than that of
FIGURE 34: Spectrum of 9kHz + 10.05kHz + 20kHz intermodulation signal.


the line input alone in Fig. 10, but the power amplifier does add its own tilt to the initial tilt of the preamp 40Hz. The square-wave response to 1kHz into 8 (not shown) was just about perfect. The leading edge of the 10kHz square wave in Fig. 36 was slightly rounded, but when I connected 2F in parallel with the 8 load (Fig. 37), there was significant ringing at 70kHz, conforming to the response peak in Fig. 24. Overall, the L53 is a nicely performing piece of equipment that incorporates a lot of features at this price point. Full disclosure here: I own a number of NAD products: a 214 stereo amplifier, a 304 integrated amplifier, and a 4155 AM/FM tuner. The L53 met or exceeded its published specifications, except for the performance of the CD section, although it could be argued my distortion measurements are within the statistical variation for independent
tests. NAD also specifies a different noise test method (EIAJ DVD tests) than I used, and I could not duplicate their test because I dont have a DVD video test disc in my collection. A comparison of the manufacturers ratings and my measured results is shown in Table 2. aX
[Watch for listening critique results of this unit in a subsequent issue. Eds.]


1. The case for more testing, Dennis Colin letter, Stereophile, June 2004, p. 11.
FIGURE 35: Square-wave response at 40Hz into 8.
TABLE 2: Measured Performance.
Parameter Power Output, Stereo Mode Frequency Response, 8W, 4 Total Harmonic Distortion IMD CCIF (19+20kHz) MIM (9+10.05+20kHz) Damping Factor Frequency Response Signal/Noise Ratio Manufacturers Rating 250W, 0.08% THD 8 20Hz 20kHz 0.5dB 0.08% 50W 1kHz 0.08% at Rated Power >to 20kHz 0.5dB >99dB, 50W 8 A-weighting >85dB, 1W 8 A-weighting 0.8mV N/S N/S Mono: 16.1dBf Stereo: 36.1dBf Mono: 0.2% Stereo: 0.5% Mono: 69dB Stereo: 64dB 40dB 30Hz 15kHz 1.5dB 1Vpp 75 negative sync Y (luminance), 1Vpp 75, negative sync, CB, CR (color), 0.7Vpp Y, 1Vpp 75, negative sync, C (color), 0.286Vpp 75 CVBS, 1Vpp 75, negative snyc, RGB, 0.63Vpp 75 2V RMS, 1kHz 0.dB, 330 2V RMS, 1kHz 0dB, 330 PAL/NTSC/AUTO 4Hz 20kHz >100dB (EIAJ) >95dB (EIAJ) 0.008% Measured Results 250W, 0.03% THD 8 10Hz - 34kHz 0.5dB, 1W 8 0.08% - 55W 8 0.011% CCIF 20Vpp 0.0079% MIM 20Vpp 73 (1kHz), 60(20kHz) 10 to 41kHz 0.5dB, 8 109dB, 50W 8 A-weighting 91dB, 1W 8 A-weighting 0.71mV unweighted 35dB 4, 36dB 8 0.12 1kHz, 8 0.13 20kHz, 8 Mono: 16dBf (subjective) Stereo: 30dBf Mono: 0.19% (400Hz) Stereo: 0.48% (400Hz) Not Tested 42dB 20Hz - 15kHz 1.4dB Not Tested Not Tested Not Tested Not Tested Line: 1.94V RMS, 1kHz 0.26dB, 460 Subwoofer: 60Hz, 110 2V RMS, 1kHz 0dB, 90k Not Tested 4Hz 20kHz 0.28dB Not Tested Noise floor 80dB (CD) 0.009% 1kHz 0dBr


FIGURE 36: Square-wave response at 10kHz into 8.
Noise Gain Amplifier Output Impedance FM Tuner Input Sensitivity FM Harmonic Distortion Signal/Noise Stereo Separation, 1kHz Frequency Response Composite Video Out (RCA jack, yellow) Component Video Out (3 RCA jacks) S-Video Output Mini-DIN, 2 4-pin SCART video out (EUR only) Analog Audio Outputs Analog Audio Inputs DVD signal CD Audio Frequency Range Signal/Noise (audio) Dynamic Range (audio) Harmonic Distortion (audio)


FIGURE 37: Square-wave response at 10kHz into 8 in parallel with 2F.


audioXpress 2005 9


Tube, Solid State, Loudspeaker Technology
Article prepared for

Iso-Bass: A Subwoofer

Isolation Transformer

By Charles Hansen

e recently purchased a wide screen LCD television, which provides three high-definition inputs for the HD cable DVR and NAD L53 DVD/ CD stereo receiver, and four more composite inputs for the VCR and other legacy equipment. Once I had all the equipment hooked up and calibrated the video with the Avia Guide to Home Theater DVD, I heard a slight hum when the VCR input was selected. It wasnt audible from my viewing position or when a tape was playing, so I let it be.
However, when I connected the NAD subwoofer line output to the subwoofer amplifier, the hum was very loud, with a pronounced 120Hz buzz. The subwoofer and TV use three-prong plugs, while the L53, VCR, and cable DVR use twoprong polarized plugs. All the equipment is plugged into the same Belkin surgeprotected outlet strip. I tried disconnecting the VCR I/O plugs, with no change in the hum. Even with the HD component video and digital audio links between the L53 and the TV disconnected, the loud hum was still there. I removed all the equipment interconnects and measured the voltages between all combinations of the equipment chassis and the input/output jack shells. There were only mV present in all cases.


I was stumped and out of tricks to remove the hum. I was not about to cir-
cumvent the two three-prong AC plugs with a cheater adapterthis is a potentially dangerous practice. I wished I still had the Jensen Iso-Max I tested for the October 2006 issue of audioXpress1,2,3. The Iso-Max is a wideband stereo line-level shielded isolation transformer unit. I didnt really need wideband because the mono NAD subwoofer output is rolled off at 100Hz4. Perhaps a small 60Hz power transformer would do the trick. The Iso-Max has a winding resistance of 2k26 at the input and 1k90 on the output. Its input impedance at 1kHz is about 47k. I dug into my collection of small power transformers and measured the DC resistance of all the windings. Because this was a low-frequency application, I wanted a high DC winding resistance to keep from loading down the low-frequency NAD subwoofer output signal. The best match was a small 115:115V 0.6VA AC line isolation transformer. I bought a few of these transformers from Fair Radio quite a few years ago (see Sources). The part designation on the transformer primary was 6K188HF and the secondary side (Photo 1) said it was shielded (probably an electrostatic or Faraday shield connected to the frame). I couldnt find anything on the Internet when I typed in the part designation. The open-circuit secondary voltage was 140V RMS with 115V on the primary. At this point I did not know whether the low-level subwoofer signal was suf-
ficient to provide the excitation needed to couple the signal to the secondary winding. Also, the secondary winding is placed directly over the primary, which is the way a power transformer is customarily designed. This provides the best magnetic flux coupling between windings, but it is also the worst case for inter-winding capacitance. I measured the capacitance between the two windings at 46pF, compared to about 120pF for the Iso-Max. I was surprised to find it that low, but the total capacitance of each winding would certainly be larger because of the high number of turns required for the high voltage windings on this small transformer core. The next step was to connect the 6K188HF between the L53 and the subwoofer amp. Voila! The hum and buzz was gone, and the subwoofer bass sounded quite good. I stuffed the transformer into a small enclosure with the input and output phono jacks isolated from the metal chassis. The schematic

PHOTO 1: 6K188HF transformer, secondary side.

audioXpress 2007 1

FIGURE 1: Transformer configurations.
of the enclosed transformer is shown in Fig. 1A.


But what if any readers want to duplicate my subwoofer isolation method? Both Digi-Key and Mouser carry 115:115V isolation transformers, although the smallest listed (Mouser) is the 15VA Triad N48X. The winding resistance will be lower than that of the 6K188HF. Another option might be an interstage transformer designed for tube amplifier use. There are also split bobbin transformers available with two 115V primaries and two lower voltage secondary windings. These are designed to provide the flexibility to connect the primaries in series for 230V operation or in parallel for 115V operation. You can connect the secondaries in parallel or series for various output voltage/current options. You may be able to use the two primaries for the subwoofer isolation function, and connect the secondaries in series to ground to provide some inter-winding shielding. The connection schematic is shown in Fig. 1B. While you may get some electrostatic shielding by this method, the transformer core may still pick up electromagnetic fields, especially if you place it near the power transformer of any other piece of equipment. A steel chassis box can help alleviate this problem, but distance is your best choice.
47k, while the transformer 20Hz output impedance is 1k33. I decided to call the packaged 6K188HF transformer the Iso-Bass, after the Jensen Iso-Max5. Photo 2 shows the transformer in its enclosure as viewed from the input side.
PHOTO 2: Iso-Bass exterior view.
Figure 2 shows the frequency and phase response of the Iso-Bass with a 0.5V RMS test signal (the sine wave oscillator in my distortion test set has a 600 output impedance). It is reasonably flat up to 1kHz, but above that it looks as though the winding and inter-winding
capacitances interact with the winding inductance to produce a 13.3kHz resonant peak that is 7.4dBv above the 60Hz 0dBv level. The secondary load resistance here is 47k. The HF response is pretty bad compared to the Iso-Max, but it is sufficiently flat for the NAD subwoofer output, which is rolled off above 100Hz (see dashed line for response). This also illustrates why vacuum tube output transformer designers carefully arrange the windings so the resonant frequency is well above the audio band. The phase shift in a mediocre transformer could produce oscillation in a wide-band feedback amplifier. Figure 3 shows the THD+N versus frequency with a 0.5V RMS input signal. The low point coincides with the peak of the probable resonance frequency in Fig. 2, which would produce the maximum impedance. Now look at the THD+N versus input signal level, shown in Fig. 4 at 20Hz and 125Hz. As you would expect, the distortion decreases with an increase in the signal level as the noise portion of the THD+N drops in relation to the rising signal. The maximum output level of my distortion test oscillator is 6.68V RMS. There is no way to saturate the 115:115V transformer, so there is no rise in the THD+N at the maximum input levels. aX


Digi-Key Corp. 701 Brooks Ave. South Thief River Falls, MN 56701-0677 1-800-344-4539


For my own amazement, I made some additional measurements on the 6K188HF transformer. At 20Hz the input impedance is 2k04, which is sufficiently high that it wouldnt terribly load the L53 subwoofer signal, whose output impedance I measured to be 110. The subwoofer amplifier input impedance is 2 audioXpress 2007
FIGURE 2: Frequency and phase response.

Fair Radio Sales 2395 St. Johns Rd. Lima, OH 45804 Mouser Electronics 958 N. Main Mansfield, TX 76063-4827 1-800-346-6873


3. Jensen ISO-MAX Products, Hansen, C., 6/00 Audio Electronics, pp. 28-30. 4. NAD L53 DVD/CD Stereo Receiver, Hansen, C., audioXpress, pp. 38-53, Dec. '05. 5. A search on the USPTO website did not turn up an Iso-Bass trademark as of the date of this submittal.
1. Inside the ISO-MAX, Hansen, C., audioXpress, pp. 44-49, Oct. 2006. 2. Inside the ISO-MAX, Hansen, C., Multi Media Manufacturer, pp. 1619, Sep./Oct. 2006.
FIGURE 3: THD+N vs. frequency.
FIGURE 4: THD+N vs. input level.

audioXpress 2007 3



PCG-SRX41P DKS-6100 MS-6378 ESD 6060 LN40C530f1H SA-W2500 Palm Pixi SA-W305 PSR-A300 Environment XL VGN-FW31M R-259 CDE-9821R Receiver R-7A53 SC-HT500 5201 PLL RM-P6 Patriots AVR 247 Siemens SL10 ML-1250 LG UV24 Voice DUO AVR 142 XV1900A-2006 108 T Vostro 1710 YP-K3AB Pocket PC Tritonextrem Smartdevice KX-TG8070E Entertainment XD Rx-d206 PX-G5300 RL34sgps B1800X PRO Series SX-KN7000 PS2KVM2 Scanjet 5590 TL-WR340G Dimension 1100 LE37A551p2R VX724 563LE-lb563t-er- Roland DR-3 Qr170 FA826 28DG16ET Bizhub 163 Manual TA-88 315-277140 DSC-T700 N VC-7413V Receiver 300 TA-F501 Nano 4G 68-8600 DL R-64ST STR-DE585 Citizen 850 DPP-FP35 FS-1010 Imageclass D880 WD-8023FB DF9921P VLS517 KX-T7050E Lexmark E240 CTK-411 EB833 24-230 JBV ZS-D55 LSC27970ST NS-SW310 Journe F700 FAX-B160 BC95XLT SC-EN35 CCD-FX510 Warmer FR-7X AVX-7300 Lexmark X83 PS-3000 CD1553B LE22B457c6H 32LG3000 GR38N11CVF VP-L900 29PT9417 12 CD 245 Mansion Uc 201B Yamaha C-4 SRU5030-87 KCA-S220A RA-930AX KX-TG2224


manuel d'instructions, Guide de l'utilisateur | Manual de instrucciones, Instrucciones de uso | Bedienungsanleitung, Bedienungsanleitung | Manual de Instruções, guia do usuário | инструкция | návod na použitie, Užívateľská príručka, návod k použití | bruksanvisningen | instrukcja, podręcznik użytkownika | kullanım kılavuzu, Kullanım | kézikönyv, használati útmutató | manuale di istruzioni, istruzioni d'uso | handleiding, gebruikershandleiding



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101