MXR m 102 dyna comp schematic

D&R Comp - A Workalike for ei ther DynaCo mp or Ross

Compressors

The lazy, sustain-for-days sound of a lot of rock guitar sounds was only partially the result of the guitar c onstruction and/or amp settings. In many cases, the guitarist was using a compressor, a kind of volume controller that varied its internal gain to keep the sound through it at about the same level no matter what the actual signal level was. The original Dynac omp compres sor, built by MXR, was a very popul ar means of ac hieving this sound. The D-comp uses the same technology as the original Dynacomp to get the same tone. A later variant of the same circuitry was the Ross Compressor. This effect varied from the Dynacomp by using a bit of additional bypassing and filtering. Otherwise, they are identical. It's possible to make a single board to do both - this is the D&R Comp.

The heart of the D&Rcomp is a variable gain amplifier built around a Operational

Q1-5 = high gain NPN silicon, such as 2N5088; will work with 2N3904 or 2N4124. Diodes 1N914 or 1N4148.

Transconductance Amplifier or OTA. The CA3080 OTA IC (U1 on the schematic) has a differential input, and a gain that

in the original but included for modern pedal practices.

is dependent on the amount of bias current supplied to its pin 5, which sets the bias current of the device, and its gain. The rest of the circuit is housekeeping, to make the OTA happy in its role of providing variable gain, or the level detecting circuit to supply

that proper bias current.

Parts not

Electro Cap Polarity

Vb Vb

D-Comp Compressor

TRIM1

Shown with True Bypass switching for the Millenium Bypass. Connect this point to the conrol pad of the MB to get an LED indicator with the same DPDT switch.

R15

R10

2 3

C10

R11

R13

R14

C11

R12

R19

R21

C13

C14

C15

Out

R17

C16

R18

R20

SUST

LEVEL

R16

How it works:

Input capacitor C1 isolates the effect's internal DC bias level from the 0Vdc level of the guitar. In the D-Comp, Q1 is biased to a reference DC level by resistor R3 to the Vb bias voltage source created by the R13/R15 divider. In the R-comp, R4 is added in series with R3 and the junction bypassed to ground by capacitor C4. Q1 supplies a lowimpedance buffered signal at its emitter. This buffered signal is routed to the inverting input (pin2) of the OTA. The noninverting input (pin 3) of the O TA is held at the same D C level as pin

2. The OTA is biased by R10 to the +9V supply and R7/R8 to ground at th e two ends of the 2K bias-balance trimmer . The trimmer serves to balance the

R-Comp Compressor

TRIM 1

Shown with True Bypass switching for the Millenium Bypass. Connect this point to the conrol pad of the MB to get an LED indicator with the same DPDT switch.

R13

C12

R10

2 3

C10

R11

R14

Not in original pedal

R15 C3

C11

R12

R19

R21

C13

C14

C15

Out

R18

SUST

R20

LEVEL

R16

C16

R17

D-Comp and R-Comp Compressors

input currents of the differential inputs and balance out any control voltage noises as the gain of the OTA changes. In the R-Comp, R6 is added in series with R10, and the junction of the two are bypassed to ground by C5.

The maximum gain of the OTA is set by the R14 load resistor attached from pin 6 to the bias voltage, and the high frequency rolloff is set by the parallel combination of R14 and C11. From pin 6, the signal goes to the base of Q2. Q2 performs two tasks - it acts as an output buffer to drive the output signal, and also as a phase inverter to help derive the DC level signal which controls the gain of the whole circuit. The output signal is tapped from the emitter of Q2, while both the emitter and collector of Q2 provide out-of-phase signals to a rectifier-filter arrangement built from Q3 an d Q4.

Th e m ost nega t i ve p ea k of t h e bot h p h a s es of t he si g n a l fr om Q2 i s cl a mp ed to groun d by the di od es at th e bases of Q3 and Q4, providing a "rectified" signal that is as large as the peak-to-peak Q2 signal. Since there are two of these wo rking on the two ou t- of-phase signals from Q2 , the signal is effectively full wave rectified . Q 3 and Q 4 inv ert; that is, higher signals cause them to pull their collectors to a lower voltage. The collectors pull this current from R18 to +9V, pulling down on the voltage of C16. In the absence of a signal, C16's v oltage rises to nearly the supply voltage. When large signals are passed through, Q3 and Q4 pull the voltage on capacitor C16 lower. The voltage on this capacitor is buffered by Q5, and in turn dr ives the bias current of the OTA through the 500K sensitivity control and the 27K resistor.

So as a result, if there is a very small or no signal at the input, Q3 and Q4 do not pull down on C16, and the voltage at the emitter of Q 5 is high; this supplies a current to the b ias pin which is determined by the v oltage at Q5's emitter and the Sustain control in series with R12. With a high voltage at Q5's emitter, the bias voltage into Pin 5 of the OTA is hi gh, and the gain of the OTA is high . As the signal level rises, Q3 and Q4 pull the voltage on the 10uF capacitor down, so the voltage across Sustain and R12 are lower, and current is reduced to the bias pin so the OTA gain goes down. Note that the sustain control can vary the current over a wide range as it varies from 0 to 500K.

This setup effectively forms a negative feedback loop which attempts to adjust the signal level at the output of Q2 so it is almost constant. When a note is first hit, it is loud and the level feedback adjusts the OTA to a low gain. As the note trails off, the feedback circuit lets the gain come up to hold the output signal almost constant until the maximum gain of the circuit is reached and the OTA can no longer keep the tiny signal up to the desired level.

The stomp sw itch on the vintage units is a DPDT switch, but not wired for true bypass . The g uitar always connecte d to the input of the effect unit. Because the input signal must go through a 10K resistor at the input to the effect, the loading may be bad enough to cause "tone sucking". I have drawn the circuit with true bypass switching to get around this. I have also shown how to wire the u nits for true b ypass with an LED status indictor; s ee the "Millenium Bypass" at GEO , h ttp://www.geo f e x.c o m for ho w to build the circuit to do th at.

Building it:

much easier, y ou will ne ed to pay attention to the di ffere nces be twee n the D- comp and R- comp v ersions. In the Parts List that follows, I h ave shaded the entries where the D-comp and R-comp version s are different. The R-comp version uses more parts, but the D-comp version needs jumpers strategically placed. It's probably a good idea to study the diff e rences in the two population diagrams before beginning building, or in fact before ordering your parts, just to be sure that you get the right set of parts.

I like to build things in a specific order:

•

While neither circuit is complicated, and the use of a printed circuit board (PCB) does make things

clean the copper side of the board to ensure easy soldering. Acetone, kitchen scrubbing powde r or just 0000 steel wool a ll d o a good job

insert and solder in any jumpers; the jumpers on this board all replace resistors, so they are on the same

0.4" spaci ng as the resistor s. ins ert r es i s tors; all t he resistor holes are on 0.4" cen ters, so you could make a pi ece of wood or pl a s tic that's

the right width to bend many of them to the right spacing at the same time; I also use a lead bender gadget that I got at Mouser. Very, very handy.

•

Bend the resistor leads slightly to hold them in the board as you insert them, and clip them off, leaving about 1/16" of lead protru ding above the copper.

•

Solder in the resistors.

•

Place the capacitors next, clenching the leads slightly and then clipping and soldering.

+ 3 hidden pages

MXR m 102 dyna comp schematic

Specifications and Main Features

Frequently Asked Questions

User Manual