Be sure to read Part 1 before continuing onto this next segment
In the last article, I forgot to mention a few other possible modes, which are hard knee/soft knee and response. After running those down, we will get into the side chain and stereo mode stuff.
Hard Knee/Soft Knee
Some compressors offer a switch-able hard knee/soft knee mode. If there is no such option on your compressor, the compressor will incorporate the hard knee circuit.
The hard knee/soft knee option controls how the amplitude curve responds at the threshold. With soft knee, when the input exceeds the threshold, the compression ratio is less at first, then increases up to the specified ratio as the input increases. With hard knee, as soon as the input signal crosses the threshold, its subject to the full amount of compression selected. Bottom line: use hard knee when you want to clamp levels down tight, and soft knee when you want a gentler compression effect. With guitar, the hard knee action will create more punch; soft knee may work well if the level variations are extreme, as the sense of dynamics will be preserved a little better.
Response determines how the C/L (compressor/limiter) reacts to input signal level changes. With peak response, the C/L monitors the instantaneous signal level. Therefore, as soon as it gets hit with a peak such as a pick transient the compression/limiting action is immediate (subject to the attack time control setting). Average response monitors the average signal level over a short time period, and uses this signal to control the gain. Average response is used more when compressing program material, such as a song youre mixing.
Side Chain Mode
Side chain jacks are available on many hardware compressors. These let you insert filters in the compressors feedback loop to restrict or add compression to a specific frequency range. This is very useful if the amp speaker cabinet has frequency suck outs or bumps that cant be totally fixed by EQing.
Plugging in the Compressor
After you've got your mixer/recorder setup as per my example in the previous column, we need to insert the compressor in the signal path. There are a few ways to do this. This set up is basically for an analog or digital mixer that has insert patch points on each input module OR has a patch bay (or outputs on the back of the mixer) that allows patching from mixer busses to recorder inputs. The digital routing is explained as well. We went through this same information in the previous article, but are repeating it here to make it easier to follow the instructions below, in the section Splitting Off The Guitar Signal To Two Modules.
On module #9:
If you're using a hard disk recorder with compressor plug-ins, simply route one to soft module #9. If you're using a digital outboard compressor and a digital mixer, and if the compressor and mixer allows for digital wire or light pipe connections, plug into the insert patch point (if available) on module #9. If there is no insert digital connection on the mixer, there may be digital connections at the module output into the digital recorder. In any case, the following will relate so simply adapt. If you're using a mixer with built in compressors on each module, simply switch in. If you want to use an outboard compressor instead, he we go: If youre using an outboard compressor, and if there is an insert output and input patch point on mixer module (module #9 in our example), patch the insert output into the outboard compressor input and patch the compressor output into the insert input. The mixer may have a switch to activate the insert. If so, switch in. Some mixers use a 1/4 Tip Ring Sleeve connector that is used for both the input and output insert patch point. If so, the other end of the patch cable will typically have two separate connectors to patch into the compressor input and output. Check your mixer manual for the insert wiring if you need to wire up such a cable. If there is no insert patch point, the mixer will surely have bus outputs and possibly direct outputs that route to the recorder inputs. Simply patch the bus or direct module mixer output used for the guitar track (bus #7 in our example or a direct output from module #9) to the compressor input and patch the compressor output into the recorder track (track #7 in our example). For this patching, the EQ will be before the compressor, which is what we want as explained next.
Positioning the EQ in the chain
If you're using a digital recorder/mixer combination, simply position the compressor in the soft path AFTER the EQ, meaning post-EQ.
The mixer will typically have a switching mode to position the insert path pre-EQ or post-EQ. If there is no switch the odds are good the EQ will be positioned after the EQ. If there is a switch, set it to post-EQ since we want the compressor to see the EQ (if EQ is being used). Splitting Off The Guitar Signal To Two Modules
OK, your signal chain from your guitar through your recorder to your compressor is set up so its time to split off the guitar signal. For stand-alone mixers and recorders, the guitar signal needs to be split off to two separate mixer modules. One leg is the typical path we have been using. The other leg needs to be picked off before the compressor.
We will now refer back to the setup in the above section, titled Plugging in the Compressor.
If you're using routing #1, and youre using a hard disk recorder with a built-in mixer that includes compressor plug-ins with a side chain option, check the manual for routing, keeping the following logic (#2 and #3 below) in mind. If using routing #2, check your digital mixer manual to find out how to split off the signal to another mixer module before the compressor input. If using routing #3, and if the mixer does not allow a module compressor side chain input, you need to use an outboard compressor and set up the patch as in steps 4, 5, and 6 below. If using routing #4, you need to split off the signal. There are two ways to do this. If your mixer has a patch bay with mults (typically 2 or more jacks that are wired together), patch the insert send into the mult. Then use another patch cord and patch back into the insert send on mixer module #9. Grab another patch cord and patch into a non-used mixer module. If you have no mixer patch bay, you need a Y cord. A Y cord is just like a mult. It has one input and two outputs to split off a signal. Plug the single end of the Y cord into the mixer module insert. Plug one of the other two connectors into the compressor. Plug the other connector into a non-used mixer module. If the mixer allows for a separate mic and line input, plug into the line input. If using routing #5, the same applies to #4. Simply get one leg to the compressor and the other to a non-used mixer module. If using routing #6, and if using a bus to route to the compressor input, also select a non-used bus. Patch the newly selected bus into a non-used mixer module. If the mixer allows for a separate mic and line input, plug into the line input. If using the mixer module direct output to route into the compressor, use a mult or Y cord as explained above.
OK, you now have the guitar path signal split off to two separate mixer modules. Keep in mind that our guitar path is through mixer module #9 and that includes the compressor in the chain. This has not changed.
In the case of the split to two mixer modules, we will now use the mixer module not yet assigned or routed to patch into the compressor side chain input.
Note: if the mixer does not allow for parametric EQ on mixer modules, you will need an outboard parametric or graphic equalizer.
If the mixer module has a direct output and was not used for the compressor input, patch into the compressor side chain input. If the mixer module does not have parametric EQ, using an outboard parametric or graphic equalizer, patch the mixer module direct output into the outboard parametric or graphic equalizer input. Patch the equalizer output into the compressor side chain input. If the mixer module direct output was used for the compressor input, use a mult or a Y cord and patch one leg to the compressor input and the other to the compressor side chain input. If the mixer module does not have parametric EQ, using an outboard parametric or graphic equalizer, patch the mixer module direct output into a mult or Y cord and patch one leg to the compressor input and the other into the outboard parametric or graphic equalizer input. Patch the equalizer output into the compressor side chain input. Another option is to assign the mixer module to an unused bus. Patch the bus output into the compressor side chain input. If the mixer module does not have parametric EQ, using an outboard parametric or graphic equalizer, patch the mixer module unused bus into the outboard parametric or graphic equalizer input. Patch the equalizer output into the compressor side chain input.
Yes, this is a whole bunch of information for a simple patch, but I have learned it is always best to cover all possibilities. To make this as clear as possible, here is a further expanation of the signal path:
Further Explanation of the Signal Path
Guitar amp mic goes into mixer module #9 (in our example) and gets split off. One leg routes to the compressor input. Compressor output either is patched into recorder track #7 (in our example) or returns to a mixer module (mixer module #9 insert) and routes to recorder track #7 input (via bus #7 or using mixer module #9s direct output). This completes the audio path.
The other leg of the split routes to a separate non-used mixer module. The mixer module routes into the compressor side chain input (via an unused bus or the mixer modules direct output). If using an outboard parametric or graphic equalizer, the mixer module routes into the outboard equalizer and the equalizer output routes to the compressor side chain input. This completes the side chain path.
If you are familiar with the frequency sweep EQ technique explained in previous parts of this series, skip down to Finding the EQ Pocket Frequency for the Compressor Side Chain. Otherwise, here are the instructions for performing a sweep EQ technique, repeated from Part 4 Tracking Clean Rhythm Sounds:
Here is a tip for EQing any sound source while searching through the frequencies. Keep the monitor volume fairly low and start with the EQ pots set at zero gain, meaning the center of the pot or the detent spot. When switching in the EQ, you should hear no change at this point. Now, while the guitar is being played, bring up the frequency volume on the mid frequency area almost all the way up and sweep the frequencies twisting the pot through its full frequency spectrum. You will find certain areas that sound good or bad for the instrument. After finding the frequency you like, pull back the frequency volume to center and then add to taste. If you are looking for an area to roll out [Editors note: Roll out is a typical studio term used when wanting to subtract EQ level at a desired frequency.] start with the frequency volume all the way down (negative EQ level) and sweep the frequencies with the frequency pot. After finding the desired spot, put back the frequency volume pot to center (off) and start rolling out until you find the best negative level spot. Try this with all the EQ areas as well. This is how the pros work with EQ.
I do not recommend doing this in the low frequency area if the sound source has serious low end (bass frequencies) to begin with, such as the bass instrument, bass drum, tom toms or the like. The reason you dont want to play around too much with the low frequency EQ settings is that you may blow your woofers if the monitor volume is too loud. When using mid- or upper-frequencies, your ears will remind you to turn down the monitor volume down while you add gain, but the low-end frequencies will not tell your ear to turn down the monitor volume until it may be too late. In the low frequency area, try adding about +6 dB for the sweep frequency checkout. By the way, note that most home stereo speakers do not go below around 60 cycles, so in most cases adding below 60 cycles is useless for instruments that have extreme low frequency information. Again, if monitoring loud, adding below 60 cycles may toast woofers.
Finding the EQ Pocket Frequency for the Compressor Side Chain
For this procedure, we will be using a parametric or graphic equalizer. In the case of a parametric equalizer, start with the Qs set to the smallest bandwidth setting.
Before we deal with the side chain EQ, you should have EQed the guitar signal to taste and set the compressor settings to taste. Again the reason for this side chain is to either bring up a speaker cabinet suck out or bring down a speaker cabinet buildup. About 99 percent of the time, the situation will be a frequency build up that was not easily dealt with using EQ, so lets deal with that issue.
Keep the following in mind if you are the guitarist and also the engineer: After playing chromatically and finding a problem area, when dialing in the EQ to an exact setting, try using one hand to play the guitar hammering-on notes [Editors note: Using a typical fret-hand hammer-on technique.] and the other hand to dial in the EQ settings. After you think you have nailed the EQ, play the guitar with both hands. Repeat the process performing slight tweaks as needed.
OK, there are two basic ways to deal with adjusting the EQ that will be used in the compressor side chain mode.
Side Chain Audio Audition Mode: Some compressors may have a side chain mode that allows the user to audition the side chain audio signal only! In this case, activate this mode. Using The 2nd Mixer Module To Audition EQ: In this case, temporarily assign the mixer module that is used for the side chain EQ to the mixer monitor stereo bus. If using an outboard equalizer and it is patched in after the mixer module (direct output or from a bus), unplug the patch into the compressor side chain input and temporally patch into a new unused mixer module and assign it to the mixer monitor stereo bus. In either case, mute mixer module #7 (recorder track return) since all we want to hear for the moment is the mixer module used to adjust EQ for the compressor side chain.
In either case, here we go.
While the guitarist plays chromatically from the open low string on up, you are looking for a frequency build up problem (most likely within the low E and A strings). When finding a loud low note, using a parametric or graphic equalizer, using the sweep frequency technique, use negative EQ to smooth out the level and make the notes sound even in that area. Keep in mind that if using a parametric equalizer, you have started with the Q setting at its minimal bandwidth. After finding the build up area, and if the build-up or suck out involves more than just one note, but instead includes of a couple of notes next to each other, you may need to slightly increase the Q (bandwidth) setting.
Note that you may have already discovered the EQ area(s) when EQing the guitar on mixer module #9 but the side chain compression will help further since too much negative EQ on mixer module #9 may affect the overall tone in the area. This is a balancing thing, meaning that while dealing with the side chain compression, you may want to add back some of the roll out on mixer module #9.
OK, lets say that the frequency is in the 200 cycle area for the first bump. Now check other areas in case there is another bump. After finding the bumps and setting the EQ, its time to get the path back to the path we had set up.
In the case of Side Chain Audio Audition Mode, on the compressor, switch out of that mode BUT make sure the compressor side chain mode is activated. The path is back to normal. In the case of Using The 2nd Mixer Module To Audition EQ on the mixer, switch the mixer module out of the mixer monitor stereo bus and un-mute mixer module #7 (recorder track return). If using an outboard equalizer and it is patched in after the mixer module (direct output or from a bus), now unplug from the temporary mixer module and patch back into the compressor side chain input. Make sure the compressor side chain mode is activated. The path is back to normal.
Regarding the example of a frequency bump at 200 cycles, you had rolled off to taste to smooth it out. We now switch from negative to positive EQ so set the 200 cycles EQ level to about 3/4th up on it s throw. The positive EQ is going to make the compressor side chain pull back (compress) the output level smoothing out the bump. The next step is to set the level to the compressor side chain to control how much level will be pulled back at the frequency bump.
While the guitarist is playing chromatically in and around the problem build up area, start bringing up the level on the mixer module that is used for the compressor side chain mode. As mentioned, the mixer module level will increase the amount of compression in the problem build up area, which is what we want. Set the mixer module level to taste to compress just the problem area without having the compressor suck back too much. The key is to make the compressor pull back the bump to make the guitar strings sound as even as possible.
If you're working on two or more frequency bumps, use the EQ frequency levels as a little mixer within the equalizer. For example, our bump is 200 cycles. Lets say that a 2nd bump is 400 cycles. If the level of the first bump is adjusted but the 2nd bump at 400 cycles is too loud, pull back the 400 cycle frequency level to taste.
Important! We are cranking gain from the parametric or graphic equalizer that will most likely add overall compression in all frequencies to a certain degree! The equalizer with have at least three separate adjustment areas so to avoid affecting the overall compression, if using a parametric equalizer, set the Q setting to wide on all other frequency settings not used. Set the frequency levels to total negative level. While the guitarist is playing chromatically in and around the problem area, experiment with frequency settings and set to areas that do not affect the build up problem, meaning stay away from that area but pull back all other areas.
For example: Set the mid band to 3 kHz and set the treble area to 10 kHz. Simply make sure that these setting do not affect the 200 cycle area, meaning with a wide Q setting, the bandwidth will be very wide. Regarding a graphic EQ, since the settings will most likely be in 3rd octaves, pull down all frequencies one band away on each side of the 200-cycle area. You may want to also pull down the band next to the 200-cycle area, if that does not affect the compression squash in the build up bump area.
Fixing a Suck Out Problem
It's best to deal with suck out problems using additive EQ (on mixer module #9) because, if you're using negative compression for a suck out, and if you're using the compressor for overall punch, the suck out note(s) will sound weak. OK, no rules. There may be times when a suck out needs the compressor help. If so, simply reverse the concept of the bump fix using the side chain EQ to push up all other frequencies while pulling back the suck out frequency.
Compressor Stereo Link Mode
The link switch in stereo compressors switches the mode of operation from dual mono to stereo. Linking the two channels together allows changes in one channel to affect the other channel, which is necessary to preserve the stereo image. This works as follows: If either channel crosses the threshold that causes compression, the other side gets compressed by the same amount whether or not it crossed the threshold. With guitar, linking matters only with stereo outputs or if you're compressing a composite guitar signal (e.g., direct sound and some amp sound).
Yes, that is the standard concept but check this out: If the compressor does not allow a side chain mode, this mode will allow the same result! In this case, instead of patching into the side chain, you patch into the non-used side of the compressor. All of the side chain stuff applies. Further, some mono compressors of the same model offer terminals for a jumper wire that ties two mono units into a stereo unit.