It's the moment you dread - the multitrack is happily playing back your vocal and guitar parts but the sequencer has stopped, and taken out the drums, bass, pads... There's a dropout on the sync track and experience tells you it's usually a fatal situation.

Such a dropout might be caused by audio on the adjacent track interfering with the sync tone, or there may be a drop in the signal level. It may even be down to the cable connections between the sequencer and tape recorder. Your usual solution is to record another timecode track, work out the start time and offset - and hope.

Another area in which timecode problems arise is when transferring sync code from one tape machine to another. This invariably distorts the square wave - ultimately to the point where a dropout occurs. Square one.

It all sounds like a job for the TCR1. Friend Chip, who brought out the K..AT ST remote controller some months ago, have recognised the need for a unit which can generate a high-quality square wave from practically any distorted waveform on tape.

The TCR1 is a full-width 1U-high rackmount unit just 3cm in depth. The front panel has controls for Sensitivity and Level along with LEDs indicating Signal and Error. It also has a couple of diagrams showing a particularly ropey input waveform followed by a precise output version (like the soap powder ad with beetroot stains). The rear has two quarter-inch jack sockets for Code In and Out and a 2.1mm socket for an external PSU (not included). If you want to be able to patch in and out on the front panel, there are holes in the chassis to accommodate extra jacks. The small size of the TCR1 belies the fact that it contains no less than eight chips on a neatly laid out, double-sided PCB.

The best way to use the TCR1 is to patch it into a patchbay so that the connections can easily be changed between use on recording SMPTE to tape, and on playing the timecode back. When recording SMPTE, the Sensitivity control is set fully to the right and then backed off anti-clockwise until the signal LED remains constantly on. The Level control is then used to ensure that the correct level of timecode is recorded to tape, which will vary between multitracks. On playback, the Sensitivity control is used in the same manner, while the Level control ensures that the Sync LED on the SMPTE unit remains constantly on. Generally, the same settings work for record and playback, which means that the TCR1 is practically "invisible" in use.

The square wave produced has edges which slope slightly; this helps to minimise the crosstalk between the track used for SMPTE and the adjacent track. While the unit doesn't "jam-sync" in that it will not continue to generate timecode if a dropout occurs, it will continue to send a data "1" signal which apparently confuses many SMPTE readers into believing that timecode is still being sent for two or three frames. Ingenious.

Results in use were very interesting. One of the principal problems with using narrow tape (cassette or quarter-inch reel-to-reel) is getting the level on tape right. Too high a level gives crosstalk; too low a level gives intermittent dropouts. It was possible to record SMPTE at -15dB on a Fostex R8 and still get a perfect lock on playback. In fact, it became clear that using the TCR1 on playback really is only half of the answer; by using it to record timecode as well, there's definitely an improvement in crosstalk for signals recorded at the same level. While I couldn't test it, I would expect the results with a four-track cassette to be as good if not better. This is because crosstalk on cassette is a serious problem and so timecodes tend to be recorded at a dangerously low level.

The idea of the two- or three-frame dropout compensation couldn't really be tested either as it is very difficult to deliberately create such a dropout. However, the delay between Code In and Out was measured with a dual digital storage oscilloscope and measured as being 140ms-200ms. Apparently, with a perfect input signal, the manufacturers quote the delay as being 125ms with a worst-case figure of 200ms, which the test agreed with. One oddity is that the waveform is inverted, but as SMPTE is a biphase signal, this is of no consequence.

Currently, no PSU is included with the TCR1; the transformers supplied with Roland Boss units work perfectly, however, as long as the polarity is reversed. Fortunately, a diode is positioned inline with the power input to protect the circuitry should you use a power supply of the wrong polarity (why don't all manufacturers do this?).

At its price the TCR1 is unreservedly recommended, especially for cassette and quarter-inch based studios. Bearing in mind that it's also perfect for transferring code from one tape to another, anyone working with audio-visual who moves audio from one system to another (quarter-inch to video, say) and also needs to copy the SMPTE would do well to have one to hand. I bought one for use with my Fostex E16 - the level on tape no longer wanders up and down like a thermometer reading.

Price £79 including VAT, plus £2.93 p&p. A power supply is available for £8.50 including VAT.

More from Q Logic, (Contact Details).