There are three required modifications on the Rev1 PCB and several optional modifications that enhance performance.

Required Modifications

1. You must cut two traces on the component side and one trace on the solder side.  The specific runs to cut are indicated in the photos.  One of the cuts is obscured by U7 and is the trace connecting the main power run to pin 14.

2. You must add three wire jumpers on the solder side as indicated in the photos.

3. Make sure you use a low leakage capacitor for C20.  I used a Mouser 647-UKL1H4R7KDDANA capacitor.

Optional Modifications

4. There is insufficient spacing between R46 and J9.  It is easier to mount R46 on the solder side.

5. The fast edges of the gate circuit induce slight glitches at the peaks of the sine and triangle waves.  You can see them on a scope but they are fast enough to not cause any artifacts.  I found that by increasing R20 to 10K and paralleling a 0.02 µF capacitor it minimized the glitches.  I simply piggybacked the capacitor on R20 and is shown in the photo.

6. The useful range of the frequency control is about -5.8 volts to +5 volts.  The control range is -15 volts to +15 volts so there is a lot of "dead" travel at each end of the control.  You can change R57 to 300K for a -5 to +5 range. You can improve the coarse control range by changing R57 to 255K and adding a 8.06K resistor in series with the CW end of the potentiometer (between the potentiometer and +15 volts on pin 1 of the coarse control connector).  This will provide a -5.8 volt to +5 volt range.

7. The delay will false trigger if the delay trigger input is unconnected.  Normalling the jack to ground will prevent false triggering if there is no cable plugged into the jack.  I added a 47K resistor from the base of Q3 to ground to eliminate the false triggering.  I could get some really interesting waveforms with this false triggering as each cycle would cause a delay trigger which would truncate the current cycle for the period of delay!

8. I often run +/- 5 volt signals into inputs such as the delay trigger.  The breakdown voltage of Q4 is rated at 6 volts so I added a protection diode from the base to ground.  This improves the margin if a negative voltage is applied to the delay trigger input.

9. I used a 2 µF timing capacitor for the 8038 which produces a frequency range of 0.5 Hz to 160 Hz.  The sine wave holds it's shape to 0.5 Hz although the amplitude drops off below 1 Hz.  The triangle and ramp hold a reasonable shape to about 1 Hz and then deteriorate and the amplitude decreases below 1 Hz.  Here is a 0.5 Hz sine wave with amplitude about +/- 3 volts.

10. I found the calibration of the volt/octave a bit tricky.  You have to let the LFO temperature stabilize.  The easiest way to adjust the trimmer is to add 1 volt to the volt/oct input and adjust the trimmer for 2X the frequency.  Unfortunately this adjustment will also affect the base frequency so you don't know the exact 2X frequency.  I found it easier to disconnected the coarse and fine controls so a 0 volts input would have 0 volts on the trimmer so it would not affect the base frequency.  This way I could measure the base frequency at 0 volt input and adjust the trimmer for 2X the frequency with a 1 volt input.  Then I could check and readjust the trimmer for 4X the frequency with a 2 volt input and 8X the frequency with a 3 volt input.