Sound Lab ULTIMATE EXPANDER Block Diagram    Table Of Contents

The Sound Lab ULTIMATE EXPANDER provides a number of stand alone modules and a flexible signal routing section for interaction and use with the Sound Lab ULTIMATE. You won't believe the added capabilities this project provides to your ULTIMATE until you build one for yourself.

State Variable VC Filter    Table Of Contents

The Sound Lab ULTIMATE EXPANDER includes a stand alone Voltage Controlled State Variable Filter. There are two signal inputs, two control voltage inputs (providing linear CV to exponential frequency response), and a resonance control voltage input. There are three outputs: Low Pass, Band Pass, and High Pass. The initial frequency control adjusts the response of the filter from sub-audio to ultra-audio. When the resonance control is advanced to maximum the VCF can be used as a sine wave oscillator. This is essentially a clone of the MFOS "State Variable VCF 12dB/Octave With VC Resonance" with some minor tweaks.

Two signal inputs can be applied to circuit points X7 and X8. The .1uF input caps into 1 meg input resistors reduces the high pass effect that a smaller input resistor would have since pin 2 of U3-A is a virtual ground. Square waves of relatively low frequency will not differentiate too badly with this input arrangement thus preserving their low frequency content. Signal levels of +/-5V are expected. If you have higher signal levels then increase the values of R24 and R30 to insure that U3-A is not clipping when you feed in your signals. The opposite would be true as well (for lower levels increase the values of R24 and R30 to get adequate signal to noise ratio). U3-A acts as an active mixer with a gain of .30 (with values shown). The output of U3-A is fed into the filter via R19 (100K resistor). The high pass output appears at the output of U3-A also and is fed to the X6 (High Pass) output via R13 2K.

This VCF uses the LM13700 dual transconductance amplifiers as voltage-controlled integrators. There are two of them in a chain and they operate in the same way. U2-B's transconductance is controlled by current flowing from ground via Q1 and R15 into U2-B pin 16. U2-B acts like a voltage-controlled resistor that in conjunction with C2 is an RC filter. Since the output of U2-B is a current the signal is actually integrated onto C2. The filtered (actually integrated) signal is buffered by U4-B and fed to the next stage (a portion is also fed back to the input of U2-B via R22). R14 is used to bias U2-B's linearizing diodes on (which is advertised to reduce distortion through the amp). R25 compensates for the positive offset applied via R14. The goal is to keep the signal path of the system as close to operating about ground as possible. In practice you will see as high as +/-200 to +/-300 mV of offset at any of U4's outputs but that's fine. If things were operating near the rails we would have a problem. Transconductance amp U2-A and associated components comprise the second integrator in the filter.

Each filter section contributes 6dB/octave filtering (thus the two together result in 12dB/octave). The cut-off frequency control voltage inputs (X4, X5 and X65) are applied to U1-A via the 100K input resistors R9, R10, and R160. Control voltage of between -5V and +10V are expected. Initial Cutoff Frequency control (R4 Linear 100K pot) is used to set the initial cut-off frequency. R5 limits the low range of voltage available at the wiper of R4 in the extreme low position. The top of the pot goes directly to +12V but the board has a jumper that is the same length as a resistor in case you use +/-15V and need to limit the high end of the C.O.F. adjustment. The summed control voltages are inverted by U1-A (inverting amp gain of .02) and fed to trimmer R1. The 20mV per volt output from U1-A can be trimmed to the requisite 18mV/volt by R1 whose wiper drives the control transistor (half of U7 SSM2210 (pins 1,2,3)) of the exponential V to I convertor. This results in a 1V/oct response in filter cut-off frequency. Current through NPN transistor U7 SSM2210 (pins 1,2,3) is mirrored by NPN transistor U7 SSM2210 (pins 6,7,8) which sinks current from the bases of Q1 and Q2. These two transistors mirror the current flowing into the collector of NPN transistor U7 SSM2210 (pins 6,7,8) and sink current from the "amp bias input" pins of the LM13700s to ground via current limiting resistors (R15 and R17). The result is that both integrators are controlled simultaneously. Since both are tuned to the same cut-off frequency very little signal above the cut-off frequency gets through the low pass portion. The low pass output is critical to the operation of the other filter modes (as described above). U3-C (and associated components) buffers and amplifies the low pass output. U3-B (and associated components) buffers and amplifies the band pass output. U3-A (and associated components) is the input mixer and doubles as the high pass output buffer. All filter outputs are protected against shorts by 2K resistors. The band pass output is is inverted and given a gain of 3 by U3-D which feeds the resonance/feedback circuit (Point RA).

To calibrate the VCF for 1V/octave response use trimmer R1 in conjunction with trimmer R2. Advance the resonance control to maximum. Adjust the initial frequency so that the filter is oscillating at about 100 Hz or so. Apply a calibrated 1V/octave keyboard voltage to one of the CV inputs so that you can play several octaves on the keyboard. As you play a series of note-octave-octave-octave adjust trimmer R1 until the sine output frequency changes by an octave as the subsequent octaves are played. Raise the initial frequency to about 500 hz and continue to check the octaves using R3 to correct the octaves as the frequency gets higher (and tends to go flat). The two controls have some interaction so you may need to go through the process more than once before you get the expected response. I have achieved a solid 4 to 5 octaves of good tracking using this method.

Point RA on Schematic Page 1 (U3-D pin 14) connects to corresponding point RA on Schematic Page 2 (to resistor R49). The other side of R49 goes to the non-inverting input of U6-A (half of a LM13700 dual transconductance amplifier) that controls the amount of negative feedback applied to the input of the filter (point RB on Schematic Page 2 connects to corresponding point RB on Schematic Page 1). Adding negative feedback results in the characteristic filter ringing that adds interesting harmonics to the original signal. You will also notice that the output signal level will decrease as the feedback is increased. Simultaneously the amplitude of the ringing will increase. This control is referred to as resonance because the harmonic content of the original signal is accentuated when the control is advanced. At maximum feedback the circuit will produce a very pure sine wave that can be used as a pitch source or control voltage. When the resonance is at about mid level and noise is used as the input signal the output will produce pitched noise. The resonance control circuit simply applies from -7.2 to +7.2 volts to R46 that controls the "amp bias input" of U6-A and thus its transconductance linearly. The resonance control signal X11 is expected to be between -5V to +5 volts or so.

The Sound Lab ULTIMATE EXPANDER includes a stand alone Voltage Controlled Amplifier. It has one input, one output and two control voltage inputs as well as an initial amplitude adjustment. It provides an exponential response to linear input voltage.

This exponential response VCA uses U6-B (half of a LM13700 dual transconductance amplifier) as the gain control cell. The input voltage applied to X15 and X16 is converted to current by the linear V to expo I section consisting of summer U8 (TL071) Q3 and Q4 and R61. This is the simplest form of V to exp I and is not temp compensated or mirrored. Since volume is not something the ear perceives nearly as precisely as pitch this works quite well and is very simple and effective. The current into the collector of Q4 drives the "amp bias input" via resistor R62. The LM13700's built in buffer is used to drive the output. The Initial Gain is used to adjust the effect of control voltage on the VCA's output. Use it in conjunction with the applied control voltage to get the desired response. Trim pot R73 is used to adjust the output offset to as close to 0V as possible with no input connected. Trim pot R69 is used to trim the input level so that the largest applied signal does not cause the VCA to clip and distort.

The Sound Lab ULTIMATE EXPANDER includes a stand alone ADSR Envelope Generator. In addition to the Attack, Decay, Sustain and Release controls there are 2 gate inputs. NOTE THAT THE GATE INPUTS ARE ALL TIED TOGETHER The gate input on the auxiliary panel and the gate inputs on the main panel are all connected. This is done so that a gate input can be connected to the banana jack or auxiliary panel and then the remaining two gate connections can be used to connect that gate to two other places. The auxiliary trigger input and the banana trigger input are also connected to one another. There is a manual gate button and a long/short range selector switch. In short range cycle times from milliseconds to seconds can be achieved. In long range cycle times from sub second to twenty or more seconds can be achieved. There are two ADSR output banana jacks for connection to the CV inputs of other modules. This is exactly the same ADSR design as the MFOS "ADSR Envelope Generator"

With gate and trigger inputs at 0V the circuit is at its quiescent state. Point R is high in this state because both inputs of U10-D are low. Point R being high causes analog switch U12-C to be on which discharges C35 to ground via the Release pot's resistance setting. When the gate input is brought slightly above 2 volts the comparator comprised of U9-A and associated components goes from -V to +V very quickly. This pushes current through C31 and drops a high going spike across R83. The spike across R83 causes the flip-flop comprised of U10-A and U10-B to output a high logic level on pin 4 of U10-B. This high level turns off analog switch U12-C and turns on analog switch U12-B which begins to charge C35 (or C35 and C36 in parallel if S2 is closed) toward V+. The voltage on C35 is sensed by U13-C which acts as a comparator with a threshold of 10 volts. When C35 charges to 10 volts U13-C's output goes from -V to +V very quickly. The positive excursion of the output of U13-C (pin 8) is fed to the input of U10-B via D6 and causes the flip-flop comprised of U10-A and U10-B to output a low logic level on pin 4 of U10-B. The presence of the gate keeps U10-D's output low but also allows U10-C's output to go high turning on U12-A which allows C35 to discharge to the level of U13-A's output (which is determined by the sustain pot setting). When the gate is released C35 is discharged to ground at the rate set by the Release pot when analog switch U12-C closes. If the gate is applied and released before the decay cycle the output of U11-A goes high and resets the flip-flop comprised of U10-A and U10-B so that the release phase is entered immediately. Application of trigger pulses (along with the gate present) after the attack phase has completed sets the flip flop and starts the attack cycle over again. Application of trigger pulses (without the gate present) results in a complete attack phase followed by the release phase.

The Sound Lab ULTIMATE EXPANDER includes a versatile signal routing section to interact with the Sound Lab ULTIMATE.

The SIGNAL MIXER is a simple 4 input summer (U14-A and associated components) with volume controls for each of the inputs. The mixer provides a gain of 1 and is menat to be used for the high level audio sources of the Sound Lab ULTIMATE and Sound Lab ULTIMATE EXPANDER. The outputs of the mixer can be routed to the line out by means of switch S3 "Line Out Select". The output of the mixer is also available via banana jacks on the main panel. Outputs in the signal routing section are protected by 3K resistors. Shorting to ground or any other input or output will not result in damage to either. U14-B is a mixer that combines the line output of the Sound Lab ULTIMATE with either the output of the SIGNAL MIXER or the VCA's output. The output of U14-B is the line output of the Sound Lab ULTIMATE EXPANDER (and the input to the headphone amplifier section). The Sound Lab ULTIMATE's line out can be mixed with the output of the EXPANDER's 4 channel mono mixer or with the output of the EXPANDER's VCA. Sources from the Sound Lab ULTIMATE (oscillators, noise, etc) can be mixed using the four channel mixer and then routed through the VCF and VCA of the EXPANDER. This adds a ton of versatility to the ULTIMATE/ULTIMATE EXPANDER combination. The level of the Sound Lab ULTIMATE's line out is set using R100 LEVEL control. The EXPANDER's line output buffer can apply up to a gain of 3 to the Sound Lab ULTIMATE's line output if necessary (dependant on the setting of R100).

The ULTIMATE MIX-BUFFER buffers the raw output of the Sound Lab ULTIMATE's mixer (see the Sound Lab ULTIMATE's "EXPAND ME" tab for information regarding adding the ULTIMATE MIXER OUT jack) and provides two banana jacks (ULTIMATE MIXER OUT) to connect the buffered ULTIMATE mix to the EXPANDER's modules.

The EXT'ERNAL' SIGNAL BUFFER section is provided to allow an external source to be input to the EXPANDER and then processed through it's modules. The section provides a gain of 100 but the input pot provides a means to reduce the amplitude of the signal appearing at the EXTERNAL SIGNAL BUFFER OUT banana jacks. This is primarily an input for electric guitar and/or a microphone either of which require sufficient gain to get them to synthesizer levels (several volts). If you desire to reduce the gain of this amplifier simply reduce the value of R120. The gain formula is 'value of R120' divided by the 'value of R119' + 1 since the op amp is non-inverting.

Wiring tips. Use tightly twisted pair or a thin flexible coax cable for audio input signal routing to prevent signal reflection in adjacent signal wires. If you don't do this you will hear bleed through between signal sources (EXT IN TO R123 and to PCB, ULTIMATE LINE OUT INPUT, ULTIMATE MIXER INPUT, LINE OUT, MIXER POTS TO PCB). Even after using twisted pair or coax you may hear a teeny tiny smidgen of a bit. If you don't use it you'll swear everything is connected to everything else. The PC board has ground pads next to signal inputs for connecting the coax shield. Remember you can strip back a bit of the coax carefully solder a wire to the shield (don't melt the inner insulation through) and then put a piece of heat shrink over that to make connecting the shield to the board more convenient. Be consistent and connect the shield on the board side or the chassis side. You don't need to connect it to both and may even contribute to ground loop noise is you do.

The CV-INVERTER does just what it's name implies. It inverts the voltage (or signal) presented to it's input and provides it at the bar OUT banana jack.

The CV-DISTRIBUTOR does just what it's name implies. It takes an input voltage (or signal) and buffers it to it's three outputs.

HEADPHONE AMP. The line out of the EXPANDER goes directly to the input of the HEADPHONE AMP. The headphone amp consists of two LM386-4 low voltage amplifiiers each of which drives a headphone in a set of stereo headphones. This is more than enough power to listen to the ULTIMATE/ULTIMATE EXPANDER combination. The bypass cap that is rarely mentioned anywhere regarding the LM386 IS mentioned in the power supply rejection spec of the chip which is specified with a bypass cap of... 10uF. This supposedly helps with power supply noise reduction. The LM386 makes a bit of hiss but it is pretty low level compared to normal listening levels. Remember you can damage your hearing just as easily with headphones as you can standing next to a jet... don't go nuts with the volume if you want to be listening to Brahms or hear your grandchildren tell you how cool you are when you get older (OK enough fatherly admonition).

ATTENUATORS. Like the attenuators of the ULTIMATE the EXPANDER provides two more so you can attenuate whatever you want to. I went a bit higher on the resistance of these to provide a bit of variety and load the sources less (especially those protected with 3K resistors).

This is as simple as it gets when it ccomes to an analog multiplier (Ring Modulator). U20 is just buffering the signals applied to the two op amps. The inputs to the AD633 definitely work better with low impedance sources. The AD633 does all of the multiplying work and provides a buffered output to boot. Whatever signals you put into the X-IN and Y-IN inputs will be multiplied (divided by 10 thanks to the chip) and presented to the output banana jack (XY-OUT). You can amp up your guitar or a microphone with the EXT'ERNAL' SIGNAL BUFFER and use it as an input (say X-IN). Put the VCF into oscillation and use the LP output as the other input (say Y-IN) resulting in no end of clanging, bonging good fun when you listen to XY-OUT.

The envelope follower has a lot of uses. The signal presented to X61 (INPUT) is full wave rectifed by U22-A and U22-B and associated diodes D8 and D9. I found that the full wave rectifier suggestions in the National Semiconductor application handbook always used the diodes in the feedback loop of the op amp but I didn't like the operation of those circuits so I went with this. This arrangement provides a more "precision" full wave rectification in my book and allows the rectified signal to reverse direction very close to ground instead of several hundred millivolts away from it. The buffered and fully rectified signal is fed to the averaging section where it is smoothed to taste via the Follower Lag control. D7 insures that the start of the envelope comes up nice and fast and R39 insures that the envelope voltage returns to ground quickly. The Follower Lag pot R150 (1M linear pot) and R149 in series dump the signal onto C72 (.22uF cap) and smooth the envelope signal further. The pot allows the follower to either snap to or lag a bit. The output can be used to gate the ULTIMATE's or the EXPANDER's envelope generators since they both have schmitt trigger inputs. The output voltage can be used to modulate the VCA, VCF, etc. etc.