While doing business as Transient Sound, my biggest advances in recorded sound quality began with direct recording via external mic preamps. My Soundcraft Spirit 16-channel 8-bus board became exclusively a line mixer for monitoring and mixdown of track outputs. During two or three sessions of extensive modifications, I simplified and optimized the mixer for this more specialized role.
I've diagrammed an input channel with modifications noted at right. See also this link for block diagrams of the channel signal path before and after modification. I modified all 16 channels in this manner. The highlights:
Dual function of channel eliminated. Each channel had a "monitor path" and a "channel path" leading to separate stereo busses. The source for each path was determined by the Chn/Mon swap switch as either the mic/line preamp or tape return input. On modification I removed this switch and hard-wired the channel for exclusively the tape return input.
EQ split switch converted to EQ in/out. As a compromise to allow EQ of both the "channel" and "monitor" paths in the stock Soundcraft channel, a switch let you hook the H and L shelving knobs to one and the H-mid and L-mid sweepable controls to the other. I converted this to simply an EQ bypass switch. This is a 3PDT switch (see "before" block diagram), and I wired the three parts of this switch in parallel as a SPDT switch with enhanced reliability.
Side chain jack eliminated. Reliability of the insert points deteriorated just a few years into this board's service. "Normal" became intermittent. I simply bypassed these jacks (along with nearly two feet of associated PCB traces per channel!), and instead used a high-quality normalled patch bay at the channel inputs. (Photo of the patch bays.)
Active components removed from the mic/line preamps. Because the channels' preamps were not being used, I removed the two transistors and one TL072 dual op amp associated with each. This lightened the load on the power supply, allowing more overhead for op amp upgrades.
Switches upgraded or replaced. I replaced all pushbutton switches assigning channels to the L-R (main) mix bus with new gold-plated units. I used one stereo group bus frequently for drum sub-mixes, so I replaced these with all new standard grade switches (arbitrarily, group 5-6). Also replaced were each channels "on" and PFL switches. Here is a photo of all the switches I pulled out.
Coupling capacitors upgraded. I replaced all electrolytic capacitors in the audio signal paths of this mixing board with new Panasonic HC or HFS radial units. I upgraded voltage ratings in most cases to provide physically more robust replacements, and bypassed all of them with 0.1 microfarad stacked polypropylene film capacitors. (In audiophile circles, film cap bypasses are believed to help electrolytics out on the high-end for a more "airy" natural sound.)
Op amps upgraded. Each channel's PCB had five TL072 dual op amps. One was in the mic/line preamp which (as noted above) I removed and did not replace. Two were involved with the EQ, and one each were for input and output. I think TL072s are fine chips and the audio world is full of them as you probably know. But I wanted to see if more expensive op amps made a difference (individually small improvements are summed in a mix). I replaced most of the remaining TL072s with different combinations of Burr-Brown/TI OPA2604 and OPA2134 packages, as well as some OP275 chips from Analog Devices. Go to this link for a table showing these upgrades. The results were generally positive according to my own subjective impressions. A significant exception was that OPA2604s are incompatible with the EQ section, causing a soft squeal at some EQ settings (see channels 1, 2 and 14 on the op amp substitution table). My op amp substitutions caused a net increase in load on the +/- 17V power supply rails. However, this was within the limits of the board's Soundcraft CPS-150 power supply. The same power supply is used with the larger 24 x 8 frame Soundcraft Spirit.
Star ground added. Taking a cue from Eddie Ciletti, I disconnected all ground connections to the ground busses in the main ribbon and instead hard-wired each of the mixer's PCB grounds individually to a common grounding block. I used a heavy-duty aluminum bar with 22 screw terminals mounted on a wood block (to locally isolate it from the mixer's chassis). I connected this block to the ground (common) pin of the power supply input jack (which in turn is connected to the mixer's steel chassis).
Master channel modification and mixdown simplification. I modified the master channel and group 5-6 modules in the same "spirit" as I did the input channels. This included switch replacements, and capacitor and op amp upgrades. But perhaps my most important "modification" was simply to change where I took the analog output: For almost all of my mixdowns, I fed the DAT machine/Mac-running-Sound Designer II from the master channel's pre-fader side-chain send (i.e., directly from the bus summing amps). This avoided two op amp stages and several electro-mechanical interface points (i.e., normalled jacks, switches, and pots). Not having use of the master fader was a small price to pay for this elegant shortening of the signal path. I executed fades with the computer during editing instead. Master faders are a relic of days past, and I recommend everyone using an analog mixer for recording mixdowns to a computer simplify their signal paths like this.
Back to Clark Huckaby Main Page
Copyright 2005 Clark Huckaby. All rights reserved. Commercial use of this content without first obtaining the expressed permission of Clark Huckaby is prohibited by law. Distribution of copies beyond personal use and/or re-posting of this page is prohibited by law.