1975 Revox A700
Revox A700
Test Report (Studio Sound and Broadcast Engineering, October 1975)
FOR MANY YEARS the name 'Revox' has been associated with high
quality recorders in the professional/semi-professional and betterendowed domestic applications. For some reason Revox have seen
little competition in the field; while others have designed recorders to
compete with Revox, their end products have either been too
expensive or lacking in the sound construction and good long-term
performance that is associated with the Revox trademark.
While I do not know the numbers of Revox type 77s sold,
considerable confidence in the sales of the new type A700 is
confirmed by the fact that the pricing of the machine is based on 100
000 units and that the development of the specialised integrated
circuits has probably cost £20 000, let alone the other development costs. With Revox working on such a
large-scale production, it is of course difficult for newcomers to compete in this rather specialised market,
and with a few exceptions the following review shows that the type A700 is indeed a first-class recorder.
Tape transport
The basis of the tape transport is a flanged alloy die-casting, which is machined for the attachment of the
three motors and the head block, itself a separate die casting. The remainder of the unit is in the form of a
cadmium-plated steel chassis to which is attached the electronic parts and the heavy mains transformer
which is, of course, better kept off the main die-casting.
Constant tension winding is obtained by servo control of the two outer-rotor type spooling motors, with the
resulting capability of handling either 267 mm NAB spools or cine centred spools; however, NAB adaptors
are not supplied with the recorder. Tension control signals are derived from tension arms adjacent to each
spool, the tension arms being spring-loaded with their position sensed by a variable transformer which
electrically controls the spooling motor torque. Proceeding towards the head block, there are two large
diameter idlers, the right-hand one driving the tape timer which is calibrated in minutes and seconds at a
tape speed of 19 cm/s, and is also used to detect tape motion—again using an rf variable transformer
technique.
The head block casting forms a plug-in unit with space for four heads, the normal complement being the
erase, record and replay heads. Azimuth adjustment is really sound, with very positive head location. The
tape guidance has been given great thought; there are four adjustable spring-loaded guides on the headblock. I am, however, a little worried about the wear properties of the guides, which appear to be un-plated
brass.
As with all respectable recorders, a flutter roller is included to reduce friction noise (scrape
flutter), but I was surprised to find that a fixed guide is included between the flutter roller and the record and
replay heads. The capstan motor is a special servo motor which is phase locked to a crystal oscillator,
offering constant speed irrespective of the incoming mains frequency at the three tape speeds of 38 cm/s, 19
cm/s and 9.5 cm/s. However, when it is required to operate at different speeds an external frequency may be
used as a reference. The capstan is of large diameter, thus easing mechanical tolerances. The pinch roller
file:///P|/9105%20werkdirectorie/Manufacturers,%20Manuals/Tapedeck/tests/9160CMREV-REV-A700.htm (1 of 9)1-11-2003 11:03:47
1975 Revox A700
operates in the trailing position by a solenoid-operated arm, which also manipulates the hinged replay head
hum screen.
All the tape transport drive functions are electrically interlocked, and tape tension is even controlled in the
start and stop modes with a resulting smooth operation without any tendency to throw loops or other
untoward habits. It is quite possible to go directly from fast wind to record without disaster—the tape stops
from fast wind, is lowered on to the heads and proceeds in the record mode!
However, not all is perfect; one shortcoming is that, in spite of the fact that tape motion is detected and the
tape end is optically detected, when the tape comes to a finish the recorder takes a considerable time to stop
the take-up spool with the result that tape can fly all over the place. This defect is caused by too much inertia
and too little friction around the tape motion-sensing roller. A further matter is that, in my opinion, the fast
wind modes are too fast, such that even 3M type 207 with its matt back does not give a very clean wind on
the take-up spool.
All function controls are illuminated push buttons, the current function(s) showing a white light in the case of
non-record functions and a red light in the case of record functions. The three tape speed selection buttons
are only illuminated when the selected tape speed has been attained, providing a safety factor when
changing speed. Other than the normal tape movement buttons, there is a non-locking 'pause' button which
operates in any mode, including fast wind and two further control modes. One of these does a fast rewind
when pressed and reverts to play when released—very useful for editing; the other button provides an 'auto’
function whereby the recorder optically senses a spliced 'window' at the ends of the tape. When the end is
reached, it rewinds and reverts to its original play or record mode at the beginning of the tape.
Access to the heads for editing is quite good, and certainly there is no problem in gaining proper access for
cleaning and de-gaussing. However, it is not possible to listen in the fast wind modes where the tape is
removed from the heads by two solenoid-operated guides. Rock and roll operation is possible in the stop
mode.
Electronics
All the electronic components are mounted on printed boards, most of which are identified with component
references as an ease to servicing. The general arrangement is that small boards plug into a mother board,
which contains a number of components; however, access to these and other components is excellent and
the many integrated circuits plug into sockets as a further aid to servicing. The overall standard of
construction is excellent and the power supplies, which can be operated from all the common mains voltages
and frequencies, are protected by some eight identified fuses. One small irritation is that the mains lead is
not of the plug-in type; however, plenty of storage space is provided for it and its plug.
On the record side comprehensive mixing facilities are provided, selection of stereo inputs being by means
of two input selector switches, each of which has two (left and right) slider-type level controls. One selector
has positions for microphones of high or low sensitivity, RIAA phono, radio or auxiliary inputs, while the
second selector provides for further microphones of high or low sensitivity, multiplay/ echo, an auxiliary input
and an off position.
The microphone inputs are balanced 6.35 mm jack sockets on the front panel and, with the exception of the
radio input, which is a five-pole DIN socket; the remaining rear panel inputs are phono sockets. A further
slider control is provided for setting master level, recording on either or both of the half tracks which are
selected by illuminated pushbuttons.
file:///P|/9105%20werkdirectorie/Manufacturers,%20Manuals/Tapedeck/tests/9160CMREV-REV-A700.htm (2 of 9)1-11-2003 11:03:47
1975 Revox A700
Two VU meters are provided for monitoring level in the before or after tape conditions and are also fitted with
peak indicator lights which have a relatively fast response. I am, however, surprised that Revox have not
gone to peak type meters.
The entire replay section is switched to before or after tape by the same switch as the level meters, and can
also be switched to the four modes of operation—stereo, left, right or mono. However, of the two available
line outputs, one is before the mode switch and the other after. These line output phono sockets and the DIN
type record/play socket are at fixed level and occur before the tone control circuits which affect the remaining
outputs.
These remaining outputs comprise a power amplifier output in the form of a DIN socket and two headphone
outputs in the form of stereo jack sockets on the front panel; one of these disenables the power amplifier
output when the phones are inserted. These outputs are controlled by a slider gain control and are affected
by the bass and treble controls which take the form of rotary switches.
In addition to the audio facilities, the power amplifier output supplies a 20V dc line for switching the mains in
the auxiliary power amplifiers made by Revox. As this socket is a standard 180° DIN socket it follows that
some caution is required not to attach other equipment to the 20V dc line. Further sockets are provided for
remote control, and for external speed control; a dummy socket being fitted to the review machine for the
future addition of slide synchronisation.
Perhaps unfortunately, there is no provision for the addition of noise reduction systems, but such an addition
should not be difficult as a customer modification.
Replay performance
Checking of the replay equalisation to the NAB standard was accomplished by means of DIN calibration
tapes manufactured by BASF which are, themselves, subject to a tolerance of
1 dB at high frequencies. It
is therefore most encouraging to report that the Revox performance was within
0.8 dB from 40 Hz to 18
kHz at a tape speed of 38 cm/s and 1.2dB from 40 Hz to 18 kHz at 19 cm/s or from 40 Hz to 16 kHz at 9.5
cm/s. Clearly the machine had been most carefully aligned, and this was confirmed by checking azimuth in
terms of relative phase between tracks.
The signal-to-noise performance in terms of reference level to noise is also excellent as is to be seen from
the following figures which show the performance of the replay amplifier system without tape but with all
motors running (see table 1).
file:///P|/9105%20werkdirectorie/Manufacturers,%20Manuals/Tapedeck/tests/9160CMREV-REV-A700.htm (3 of 9)1-11-2003 11:03:47
1975 Revox A700
On the other end of the scale, it was found that amplifier clipping, at 1 kHz at a nominal tape speed of 38 cm/
s, did not occur until the fluxivity at the replay head was increased to some +15 dB above 320 nW/m, which
allows in the order of 5 dB margin for any current tape types.
Record replay performance
As with the replay only frequency response, the record/replay response with 3M type 207magnetic tape was
really excellent. Fig. 1 shows that the overall response at 38 cm/s is within
1.3 dB from 20 Hz to 20 kHz
including the minor bass boost at 30 Hz, which is probably due to head polepiece effects. It should also be
noted that the balance between the two channels is really excellent.
file:///P|/9105%20werkdirectorie/Manufacturers,%20Manuals/Tapedeck/tests/9160CMREV-REV-A700.htm (4 of 9)1-11-2003 11:03:47
1975 Revox A700
The general pat-tern of the response at 38 cm/s was followed at two lower tape speeds as shown by the
following (see table 2).
The following table shows the reference level to noise ratio for 3M type 207 tape in its bulk--erased form, and
when it has been erased and recorded with bias by the machine with all inputs shut (see table 3).
Comparison of the above table with the replay only figures shows that there is a-considerable margin
between the inherent machine replay noise and the noise from a modern tape such as 3M type 207, it is
however felt that a small improvement is still possible in the noise introduced by the record process.
Distortion at the above reference levels was less than 0.8 % third harmonic at the two upper tape speeds,
and less than 1.0% third harmonic at a tape speed of 9.5 cm/s. Three per cent third harmonic distortion
occurred at +5 dB above reference level at the two higher tape speeds and 7 dB above 250 nW/m at 9.5 cm/
s.
Intermodulation distortion was measured at a tape speed of 38 cm/s by the SMPTE method using 50 Hz and
7 kHz tones in a 4:1 amplitude ratio at the following equivalent peak sine wave levels:
While the figure at reference level demonstrates the shortcoming of the bass boost in NAB equalisation, the
figures at lower levels are good by tape recorder standards. Of similar interest to intermodulation distortion is
the effect of friction noise (scrape flutter). Fig. 2 is a narrow band analysis of a replayed 10 500 Hz tone that
file:///P|/9105%20werkdirectorie/Manufacturers,%20Manuals/Tapedeck/tests/9160CMREV-REV-A700.htm (5 of 9)1-11-2003 11:03:47
1975 Revox A700
had been recorded at 38 cm/s. While flutter sidebands can be clearly identified at approximately 10 Hz and
multiples thereof from the carrier frequency, this performance is good in comparison to very many recorders.
Of further interest is the phase relation between the two stereo tracks; this property is shown in fig. 3, which
is an oscillogram of the phase shift when recording and replaying a 10 kHz signal at 38 cm/s. This
oscillogram, which shows a maximum phase deviation of
6°, was obtained by using a Bruel & Kjaer 2971
phasemeter which has a response time of 2400°/s at audio frequencies. It is interesting to note that the
periodicity of the phase shifts may be correlated with the pinch roller diameter.
file:///P|/9105%20werkdirectorie/Manufacturers,%20Manuals/Tapedeck/tests/9160CMREV-REV-A700.htm (6 of 9)1-11-2003 11:03:47
1975 Revox A700
The investigation into crosstalk performance yielded very good results with the measured record/replay
crosstalk between channels being below -70 dB at mid frequencies, including head contour effects which
remained below -53 dB at any audio frequency, 35 Hz being the worst case.
Erasure performance was measured by recording a 1 kHz tone at 38 cm/s at the reference level of 320 nW/
m, and then measuring the residual signal after a further pass over the record and erase heads with the input
faders shut... the resulting signal was reduced some 83 dB on the worst channel, the other channel offering
some 87 dB erasure no complaints in this department.
Wow, flutter and speed
Wow and flutter to the DIN weighted method (as now agreed by the IEEE and ANSI) was measured at the
beginning, middle and end of both 178 mm cine type spools and 267 mm NAB spools at the three tape
speeds. While there was little variation of wow and flutter within a spool, the initial measurements at the tape
speed of 38 cm/s were somewhat higher than those shown in the following table, which are average values:
These results, which are well within specification, are outstandingly good for this class of recorder, and the
initial measurements were within specification. However, no reason was found to explain the higher initial
measurements, which were consistent over a period of time. The measurement of the tape speed indicated
the identical error of +0.3 % at all speeds, which is on the tolerance limit of the calibration tape used;
furthermore, the speed variation from one end of a NAB reel to the other was within
0.01 %.
Inputs and outputs
All the above performance figures were measured between the line inputs and the line outputs, the
alternative facilities were therefore investigated. So far as frequency response is concerned, the microphone
inputs to all intents and purposes were flat from 100 Hz to 20 kHz, but showed some roll-off in the bass with
a -3 dB point at 35 Hz, which is no bad thing. On the other hand the phono input was within
0.5 dB of the
RIAA equalisation curve from 20 Hz to 20 kHz.
While noise in the record section was generally good, the microphone inputs were rather excessively noisy,
the equivalent input noise being -113 dBm when loaded into 200 ohms this figure could be considerably
improved.
The sensitivities and impedances of the inputs and outputs were all checked and found to be to specification,
although some comment is appropriate. Firstly, the input impedance at the microphone input is effectively
around 6k ohms which is far higher than is required for 200 ohm microphones which are probably the most
common impedance. The measured noise from the microphone inputs correlates very closely with the
unnecessarily high input impedance, and the provision of an impedance in the order of 1k ohms would make
a very substantial reduction in the noise from this source.
file:///P|/9105%20werkdirectorie/Manufacturers,%20Manuals/Tapedeck/tests/9160CMREV-REV-A700.htm (7 of 9)1-11-2003 11:03:47
1975 Revox A700
Secondly, the impedance of the line outputs at 4.4k ohms (as measured) is unduly high for professional
applications, and while the alternative headphones and power amplifier output are available, they are after
the level control and tone controls.
On the subject of the tone controls, these are seven-position rotary switches giving 2 dB steps at 80 Hz and
10 kHz, the good characteristics of which are shown in fig. 4, which shows the response at alternate
positions.
The level meters
While most manufacturers claim to fit vu meters, inspection often reveals that the so-called meters are
nothing like the ASA standard meter not so with Revox, the vu meters are to the proper specification with
average rectifiers and the correct rise time the return time is however measured at nearer 400 ms than the
recommended time which is close to 300 ms. I am not however at all happy about the tape flux level
corresponding to 0 vu which was found to be -1 dB relative to 320 nW/m which only leaves a 6 dB margin
between 0 vu and 3% third harmonic distortion as opposed to the NAB recommendation of 8/10 dB.
On the other hand the peak indicator lamps operated at the 3% distortion point within 10 ms of an overload,
and stayed on for 130 ms thus easing readability.
Summary
During the course of this review I have been critical of any minor shortcomings of the Revox A700 for the
simple reason that this is a very interesting machine which will, without doubt, sell in great numbers; it is a
very fine machine for its price and thoroughly recommended.
I would however suggest that the microphone input impedances should be modified to reduce the rather high
noise level that results from such a high input impedance and that some attention should be paid to the zero
level settings of the vu meters.
On the credit side there are many excellent aspects of the machine; among which signal-to-noise in the
file:///P|/9105%20werkdirectorie/Manufacturers,%20Manuals/Tapedeck/tests/9160CMREV-REV-A700.htm (8 of 9)1-11-2003 11:03:47
1975 Revox A700
replay department, frequency response, speed stability and wow and flutter deserve special mention.
Furthermore, the standard of both mechanical and electrical construction are beyond reproach. In fact, the
two latter aspects of the machine put some recorders costing very much more into disgrace.
by Hugh Ford
file:///P|/9105%20werkdirectorie/Manufacturers,%20Manuals/Tapedeck/tests/9160CMREV-REV-A700.htm (9 of 9)1-11-2003 11:03:47
Loading...