UREI 920-16 Cooper Time Cube
Two grey boxes, four tubes, and the most peculiar way anyone ever faked a room.
Overview
The Cooper Time Cube is not an echo chamber, a plate or a tape machine. It is a pair of acoustical delay lines - garden-hose-style tubes with a driver transducer at one end and a pickup at the other. Somebody at UREI decided the missing ingredient in artificial reverberation was a little honest air time. The delay line housing measures 24 inches by 24 inches by 9 inches and weighs 36 pounds. The electronics chassis is a 19-inch rack unit, 3½ inches tall and 7 inches deep behind the panel, and weighs 12 pounds. Shipping weight for the whole system, cables included, is 62 pounds in a 330-pound test carton with shock-absorbing foam corners. The manual says the packaging was, in UREI's own words, "torture tested."
The idea, as laid out in the instruction book, is simple. A good live room gives you two things: a decay envelope, and, before the decay even starts, the early reflections that tell your ear how big the space is. Plates, springs and chambers can fake the decay. They cannot fake the gap between the direct sound and the first reflection. Tape delay had been the standard workaround, but the manual is blunt about its problems. A professional tape machine is bulky and expensive for what amounts to a non-productive job, and unless you run absurd capstan speeds, the distance between the record and playback heads gives you too much delay. The 920-16 gets around all of that with a physical transmission line.
Channel A delivers 16 milliseconds. Channel B delivers 14. Cascade them and you get 30 milliseconds, which the manual says approximates the first-order reflection you would hear from a wall about 40 feet away. The manual lists wow and flutter as 0%, a spec you can only claim when there is no moving tape anywhere in the signal path.
Frequency response is quoted at ±2 dB from 40 Hz to 10 kHz, typically ±1.5 dB. Total harmonic distortion stays under 1% - typically under 0.5% - at all program VU levels up to +4 dBm output. UREI notes that distortion does not increase at low levels, and warns against measuring it at full output because of high-frequency pre-emphasis in the send sections. Signal-to-noise is greater than 70 dB over a 15.7 kHz noise bandwidth. Input sensitivity spans −20 to +20 dBm for +4 dBm output. Both inputs and outputs are 600-ohm transformer-isolated and floating, so the unit works with balanced or unbalanced gear. Channel separation exceeds 40 dB. There are two inputs, two outputs, and only three front-panel controls: Input A gain, Input B gain, and a meter transfer switch. The VU meter reads +4 dBm at 0 into a 600-ohm load and can be flipped between the two outputs.
The electronics chassis carries four plug-in Universal Audio 1109 card amplifiers, two send and two receive. With the access door off, the manual gives their order from left to right: A receive, B receive, A send, B send. All impedance strapping and equalization happen on the motherboard, not the cards, so any 1109 can be swapped into any slot in a pinch. UREI recommends keeping a spare 1109 on hand as a backup for several 920-16s, which suggests these were deployed in clusters, in working rooms, where downtime cost money. The power supply is zener/transistor regulated, delivers +30 VDC to the four cards, and is switchable between 110-120 VAC and 220-240 VAC with a recessed slide switch on the rear panel. The driver transducers are lightly fused - 1/8A for 110-120 VAC operation, 1/16A for 220-240 VAC. The manual warns that sustained program levels well above 0 VU, or feeding the unit the output of a tape machine being rewound with the tape still on the heads, can pop that fuse. Do not go heavier on the fuse. UREI says so in capitals.
The delay line housing is relatively immune to external ambient sound. The manual cites tests showing 80 dB SPL at the enclosure producing less than −60 dBm at the outputs. It still needs to live somewhere sensible. The electronics chassis should not be racked directly above power amplifiers or power supplies, and should be kept away from large transformers and motors, particularly around the left-hand section where the four 1109 cards sit. The two 25-foot XLR-3 microphone cables that connect chassis to housing are furnished. Longer standard mic cables work fine up to about 150 feet, beyond which shunt capacitance starts nibbling at the top end. The manual also addresses the ground-loop buzz that anyone racking this thing in a metal rack alongside grounded audio commons will eventually meet, and suggests a three-to-two-wire AC adaptor as a fix. That advice would raise eyebrows today but was standard practice in the period.
The 920-16 was designed to work alongside UREI's own outboard gear. The manual notes that it can be paired with the Universal Audio Envelopmental Equalizer Model 509 with no change in operating parameters, except that voltage gain becomes fixed at 45 dB from a terminated 600-ohm source, 50 dB unterminated, 51 dB terminated at 150 ohms, and 56 dB unterminated at 150 ohms. That is a very specific integration path, and it suggests the Time Cube was never meant to stand alone in a chain.
The studio press of the day left a thin record. DB-Magazine (October 1972, p.15), in a piece explaining how the Dolby noise-reduction system succeeded where compressor/expanders had failed, invoked the Cooper Time Cube by name as a reference point in the discussion of what noise reduction could and could not do without one. That passing mention places the unit in the working vocabulary of American studio engineers by late 1972. Recording-Engineer (February 1972, p.4) ran a letter from Woody Langley at KTTV in Hollywood in its "Letters and Late News" column, and the June 1972 issue (p.6) carried correspondence from Bill Lazarus at Paramount Recording in Hollywood. That suggests the Time Cube was circulating among West Coast broadcast and recording facilities in its early years. None of these are full reviews, and none of them give measurements. The Cooper Time Cube seems to have been a product working engineers knew about and used, rather than one that got lab-bench treatment in the consumer magazines.
The 920-16 is a strange object - a delay line you can open up and look at, with no tape, no springs, and no digital anything. The 1109 amplifiers are standard Universal Audio cards, which helps on the service side. The manual's maintenance section is written for a competent technician rather than a factory return: check the interconnect cables first, then swap the send and receive amplifiers between channels to isolate a fault. Electrolytic capacitors and diodes in the power supply are field-replaceable. The driver transducers are the wild card. They are the parts most likely to be irreplaceable, and the fuse warning exists because they are fragile. Treat any unit with a blown or bypassed fuse with suspicion.
Original pricing, production years, and current market values are not documented in the sources available here, and this page will not guess at them. What can be said is that the Cooper Time Cube occupies a small, specific niche: it is not a reverb, it is the thing you put in front of a reverb so the reverb sounds like a room instead of a device. Thirty milliseconds of honest, physical delay, delivered by a pair of tubes in a grey plywood box. There has never been much else quite like it.
eBay Listings
As an eBay Partner, we earn from qualifying purchases. This helps support our independent vintage technology research.