Magazine Archive

Home -> Magazines -> Issues -> Articles in this issue -> View

Noise reduction explored

Studio Sound Techniques

Article from Home & Studio Recording, September 1983

This month we explore that often misunderstood link in the recording chain — noise reduction. Used with care noise reduction systems can help improve recording quality and dynamic range. Here we explain the operational principles of two of the best known systems — Dolby A and dbx.

Noise Reduction Systems

There are many different types of noise reduction systems which may be interposed between the mixing console and tape machine. In general they use much the same technique to obtain the perceived reduction in noise; compression on the record side and expansion on playback. Since compression and expansion are new terms in this series it will be time well spent if we review them.


Figure 16 shows the input to and output from a typical 'effects box' compressor. As you can see below the threshold point or where the compression ratio is set to 1:1 the compressor has little effect on the signal. Where the ratio is 2, 4 or 10 to 1 though an increase in input above threshold results in a smaller than expected increase at the output.

The compression ratios refer to the decibel levels above threshold at the input and output respectively. So for instance at 10:1 a 10dB rise (above threshold) at the input results in a 1dB rise at the output. It may appear at first sight that we are losing a lot (9dB) of signal here. In fact, because we have squashed the dynamic range somewhat we would find that the average level is increased(!) rather than reduced. If this seems confusing think of it as, the quieter signals being able to be 9dB louder, without the peaks distorting as they would have done without the compression. This is good news for the record side of the tape machine which has a restricted dynamic range from the noise point of view and a tendency to distort 'peaky' signals.


To restore our original dynamic range on playback we would need to process the signal in the reverse way to the compressor ie. expand it. The expander is set to increase its gain above threshold by the same amount as the compressor reduced it. So following the above example a 1dB rise at the input of the expander (above threshold) results in a 10dB rise at the output. So where did we lose any noise? The answer to this question is that in the above system we would not lose any noise in the area that matters most ie. the quiet passages. And this is because we did not alter the 'gain' of either the compressor or the expander in that area. This is where noise reduction compression/expansion systems differ from the type you get in normal effects units.

Low level signals

Different noise reduction systems treat the low level signals in different ways. Dolby A for example splits the frequency spectrum into four bands and compresses everything in each band by 1.5:1 once it drops below a predetermined level. Above this level no compression takes place (ratio 1:1) as by definition the presence of the signal will mask any noise underneath it.

Dbx on the other hand linearly compresses all signals by 2:1 and adds an amount of pre-emphasis (high frequency boost). The dynamic chart in Figure 17 shows the theory of tape noise reduction. Column 1 represents the dynamic range of the signal to be recorded as it leaves the desk group output. Note that the lowest part of the signal is 10dB less than the tape noise in column 3! Which means that if we recorded without noise reduction the quieter passages would be masked by 10dB of tape noise. Compare with column 5.

Instead we compress by 2:1 (column 2) squashing our original dynamic range of 60dB to 30dB and we elevate the signal so that the loudest part is equal to 0 VU on the tape machine (column 3).

Source noise

Note: that the microphone and desk noise (source noise) is still lower than the tape noise. If the source noise was higher than the tape noise it would be regarded by the machine and noise reduction system as wanted signal and treated accordingly! Remember also that noise reduction systems only reject noise originating in the tape machine.

Figure 16. Transfer characteristics of a typical compressor.

The signal is then recorded onto the machine with a clear 20dB between the lowest sound and the tape noise (3). To restore the signal to its original dynamic range it is attenuated (4) and then expanded by 1:2 (5). It is in this part of the process that the machine noise is reduced because the expander derives its overall gain from its input signal. Consequently, when the signal to the expander is just tape noise the output from it is extremely low (-100dB).

In practice it is possible with care to achieve a 90dB dynamic range using dbx and about 75dB using Dolby A. Here are my brief personal opinions on these systems.

Dolby A

1. Uses peak detectors which can cause mistracking.

2. Reduces noise by 10dB.

3. Effective on mid range sounds eg. guitar, piano etc.

4. Many engineers prefer not to add Dolby to stereo mix downs, drums, lead vocals, percussion; while others will use the record processing to compress the signal and then bypass the playback processing to obtain a brighter sound.

5. Tape recorder 'level' alignment important.

6. Standby noise from 16 to 24 Dolby's can be a problem on mix down.

7. Generally considered more subtle or sweeter sounding than other systems.

Figure 17. Principle of tape noise reduction.

8. Preferred by many for multitrack orchestral work.

9. Some types have electronically balanced inputs which not being isolated, can interact badly with electronically balanced outputs.

10. The most widely used and accepted system, with good overall performance.


1. Uses RMS detectors which gives good tracking.

2. Reduces noise by 30dB.

3. Any machine frequency response errors are doubled on playback.

4. Many engineers prefer not to use dbx on stereo mix downs.

5. Modulation noise can be a problem on continuous bass notes.

6. Standby noise from 16 to 24 dbx units is very low. Good for music with a wide dynamic range.

7. Harder sound than Dolby.

8. Compatibility can be a problem if moving tapes from one studio to another.

More from these topics

Browse by Topic:

Effects Processing


Previous Article in this issue

Tascam 234 Syncaset

Next article in this issue

On The 8-Track - Sackville Sound

Publisher: Home & Studio Recording - Music Maker Publications (UK), Future Publishing.

The current copyright owner/s of this content may differ from the originally published copyright notice.
More details on copyright ownership...


Home & Studio Recording - Sep 1983

Feature by P.A. Becque

Previous article in this issue:

> Tascam 234 Syncaset

Next article in this issue:

> On The 8-Track - Sackville S...

Help Support The Things You Love

mu:zines is the result of thousands of hours of effort, and will require many thousands more going forward to reach our goals of getting all this content online.

If you value this resource, you can support this project - it really helps!

Donations for April 2024
Issues donated this month: 0

New issues that have been donated or scanned for us this month.

Funds donated this month: £7.00

All donations and support are gratefully appreciated - thank you.

Magazines Needed - Can You Help?

Do you have any of these magazine issues?

> See all issues we need

If so, and you can donate, lend or scan them to help complete our archive, please get in touch via the Contribute page - thanks!

Please Contribute to mu:zines by supplying magazines, scanning or donating funds. Thanks!

Monetary donations go towards site running costs, and the occasional coffee for me if there's anything left over!

Small Print

Terms of usePrivacy