Classic studio signal processors such as the vintage Urei 1176 limiting amplifier impart a unique sound while keeping levels under control, but are rare and expensive. Purple Audio have an answer in the form of the MC76 -- an old idea revisited. Hugh Robjohns does the time warp again...
Some of the most useful pieces of equipment in any studio -- after the microphones, console, monitors and recorders, of course -- are the dynamics processors. Virtually every audio manufacturer in the world has at least one compressor/limiter in its product listing and there are many, many fine dynamics processors out there. Yet I would suggest that every professional studio worth its title is likely to have at least one classic Urei 1176LN limiting amplifier in its rack, as well as some hi-tech devices of the modern age. These ubiquitous devices have remained firm favourites with recording engineers across the world because of their ability to enhance almost any signal source in a fascinating variety of ways, ranging from subtle and delicate to hard and raunchy. They also happen to have a fantastically simple, but very effective, user interface which encourages the user to listen rather than just dial in the numbers.
PURPLE AUDIO MC76 £1416
A classic 1176 in a bright-purple box.
Sounds great on most sources.
Messy stereo linking system.
Not entirely intuitive to set up.
Reinvented and affordable classic limiting amplifier which is easy, if unusual, to operate and sounds fantastic on a wide range of sources.
Pre-owned Ureis of any vintage can command substantial prices these days, so a lot of people were very pleased when Purple Audio set themselves the challenge of revisiting this particular piece of audio history. Their MC76 is based very closely on the 1176LN-E variant, which the company claims to have been the best-sounding version. The product name is as much a reinvention as the machine itself because, depending on who you believe, the letters either denote 'Mono Compressor' or provide the missing two digits from the original Urei device name -- but in Roman form!
The Ins And Outs
Intentionally, the MC76 looks (apart from the striking purple paint job) extremely similar to the Urei original. Even the Purple logo bears a strong resemblance to the old Urei badge. The 2U rackmount box is about eight inches deep and weighs roughly 5.5kg.
On the left of the front panel are two large knobs for setting input and output levels respectively, whilst the smaller attack and release time controls are arranged vertically in the centre of the machine. Grouped around the large illuminated VU meter are two rows of four chunky push-buttons, the set on the left selecting the compression ratio and those on the right determining the metering mode. Between the input and output controls a small hole in the panel provides access to a trimmer for calibrating the zero point of the gain-reduction meter.
The original 1176 could fairly be described as a 'quirky' device to use. It simply doesn't operate in quite the way that you might anticipate if you have experience of virtually any other dynamics processor -- and yes, the MC76 retains all of its 'lovable' features! There is no threshold control, for example -- the input level control determines how much of But there are a couple of things to remember. The simple circuit design means that the attack and release time constants interact across the two machines and are effectively paralleled, so that the fastest attack time is slowed to double the original value. The best way to use the machines in this linked condition is to match ratio, output-level and input-level settings on both machines, and to position the attack and release controls of one unit fully anti-clockwise (slowest position). This effectively makes these controls redundant, and the controls on the other machine then determine the time constants for both. Gain reduction should track fairly accurately over the first 10dB or so, but is likely to mistrack beyond that because of variations in the transconductance values of the FETs in the two units.
Double Trouble: Linking Two MC76s
The linking procedure involves delving around the back of the machine to plug a standard jack-jack cable between the 'offset' jack of one unit and the 'direct' jack of the other. Assuming both units are switched out of circuit and are without input signals at this stage, the gain-reduction meters will probably take off in opposite directions from their resting positions at the zero mark, because the FET biasing voltages are not yet matched. The next step is to recalibrate this biasing voltage (on the machine connected with the 'offset' socket) using a trimmer pot behind a small hole in the rear panel. This process is very difficult, as you can't easily see what you're doing to the meter displays when crouched at the back of a 19-inch rack. Once the biasing has been balanced and the gain-reduction meters have been guided back to their zero points, the units can be used in their stereo-linked mode.
But there are a couple of things to remember. The simple circuit design means that the attack and release time constants interact across the two machines and are effectively paralleled, so that the fastest attack time is slowed to double the original value. The best way to use the machines in this linked condition is to match ratio, output-level and input-level settings on both machines, and to position the attack and release controls of one unit fully anti-clockwise (slowest position). This effectively makes these controls redundant, and the controls on the other machine then determine the time constants for both. Gain reduction should track fairly accurately over the first 10dB or so, but is likely to mistrack beyond that because of variations in the transconductance values of the FETs in the two units.
For the record, the attack time is variable between 20 and 800 microseconds, and release between 5 milliseconds and 1.1 seconds. The four ratio buttons are marked as 20:1, 12:1, 8:1 and 4:1, although they can be used in combination, with sometimes effective, if rather unpredictable results. Although there is no dedicated threshold control, the actual threshold point is sensibly related to the selected ratio, such that the higher the ratio, the higher the threshold. For example, at 20:1 the threshold is set to -24dBu, while a more gentle ratio of 4:1 starts with a lower threshold of -30dBu. Setting the unit up thus requires you to select the desired ratio, set the peak output level with the relevant control, and then adjust the input level to provide the required degree of 'squash'.
The four buttons to the right of the meter are configured in an equally unusual way. The bottom button switches the unit off, while pressing any of the other three powers the machine. The top button switches the VU meter to display the amount of gain reduction occurring; this is the normal operating mode. The two middle buttons show the output level from the machine, referenced to either +4 or +8dBu, although these hang the meter directly across the output without the aid of a buffer circuit, so can add distortion to the signal. These options are really only intended for initial setting of the machine's gain structure.
Few engineers would ever bother to use multiple linked Ureis for stereo or multi-channel working, simply because of the impractical method required to link the machines. A special interface box (the 1176SA) was needed. This featured an FET biasing battery, and the control voltages from each unit had to be trimmed and matched by hand. The MC76 improves this situation slightly, in that the interface unit has been built in, although there is still a need to match and trim the control voltages via a rear-panel trimmer and associated DC bias polarity switch. You wouldn't necessarily know it, but there's a lithium battery (AA-sized) tucked away inside the machine to provide the necessary biasing too. I recall a discussion in late 1997 (shortly after the MC76 was launched) regarding whether this consumable item could be replaced with a permanent biasing supply, courtesy of an extra tapping on the mains transformer, but unfortunately it appears that this idea has not been put into practice.
The MC76's rear panel reveals further improvements over the 1176, in the adoption of proper transformer-balanced and floating XLR connectors for audio input and output, and quarter-inch jack sockets for stereo linking. The original used terrible solder-tag connections on screw blocks for the audio I/O, and phonos for the linking facilities. The remaining connectors on the rear panel are an IEC mains connector, with integral fuse holder and voltage selector, plus a pair of sturdy binding-posts to link signal ground and mains earth.
Internally, the MC76 retains much of the electronic design approach of the 1176LN-E on which it is based, even down to identical component types in many instances. The balanced constant-impedance input attenuator is the same Allen & Bradley device used in the original 1176, for example, as is the UTC input transformer which follows. However, there are also a number of small but significant improvements, including tweaks to the overall circuit design and improvements to the circuit board layout. Alternative components have also been used in a few areas where there were advantages in consistency or sourcing, including a custom-made output transformer. Structurally, the Purple Audio unit is also much stronger and generally better built than the old Ureis.
The input impedance is constant at a lowish 600(omega) (at all gain settings), and a signal level of up to +30dBu can be tolerated. Maximum through-gain is an impressive 45dB, and the Class-A output stage can drive a 600(omega) load to +24dBu or a high-impedance bridging load to +30dBu. Despite the transformers on input and output, the machine delivers an overall frequency response quoted as 15Hz-80kHz, +/-1 dB. Distortion is specified as less than 0.5 percent during limiting (assuming sensible attack times) and the signal-to-noise ratio as better than 81dB (unweighted).
It has been a while since I used an 1176 in earnest, but the MC76 seems to behave very much as I remember the original, even down to the wacky results obtained by pressing multiple ratio buttons. Other reviewers have suggested
"The original 1176 could fairly be described as a 'quirky' device to use -- and yes, the MC76 retains all of its 'lovable' features!"
The MC76 certainly imparts an almost valve-like warming effect to signals with just the gentlest amounts of compression, and it works particularly well on strong vocals, most drum tracks and, especially, bass guitars -- just as the original did. At low ratios, with only about 4dB of gain reduction, the unit is virtually transparent in the way it processes the dynamic range of a signal, yet it still seems to be able to enhance the sound passing through it in an almost intangible, yet desirable way. With harder compression it produces a more raunchy, busy sound, sometimes tending towards a buzzy harshness which I don't remember from the original.
Unfortunately, linking two units together is almost as much of a pain as it was with the original Ureis, I'm sorry to say (see the 'Double Trouble' box for details), and I'm not sure many engineers would be bothered to try it. The MC76, like its antecedents, is a wonderful mono compressor but a finicky and largely impractical stereo one!
To sum up, the MC76 is a lovely machine with, as far as I can tell, all the desirable attributes of the classic 1176s in a slightly 'ruggedised' and modestly improved package. Easy to use, great sounding and offering worthwhile value for money, this device is a certain winner.