Six years after I built the first one, I decided to build a second Dannysound EN129 Thru-Zero Oscillator. The design originated in the Electronotes newsletter #129, published by Bernie Hutchins in 1982. I was a subscriber to Electronotes and had already built some circuits published there by 1982. But I never had built the Through-Zero oscillator. I’ve always liked this VCO for its dependable classic waveforms and through-zero FM capability. I’d been looking at so-called “complex” oscillators, which typically contain two similar VCOs, adding cross-modulation and wave folding. Complex oscillators are among the most expensive and reflect the design ideas of the creator. I generally like to make my own complex patches from basic modules. I have plenty of VCA and Wave Shaper modules already. Thus I came to the conclusion to get a second of my favorite analog VCO in Eurorack.

Construction

I bought the kit from Thonk. This oscillator is DIY only, and Thonk sells only full kits. This is a blessing, because some of the parts have become hard to obtain. For example the 79L09 minus 9 volt regulator is out of production. I’ve been using DI79L09ZAB instead. But the kit had a genuine 79L09. Similar for the CA3140 and CA3080. It’s a rather expensive kit, but you get a rather nice oscillator! Kit includes the machined pin IC sockets too, which I used.

Assembly is pretty straightforward. The main PCB is still version 1.0. It’s the panel PCB that’s updated to include the soft sync input and circuitry (which I’ll be demonstrating). This time I really scrutinized the schematics and found two minor differences. R214, which sets the current for the LED in the LFO pushbutton is a 220 ohm, rather than the 1K shown on the schematic. And, weirdly, there is actually no digital ground, DGND, as shown on the schematic. Everything shares the same ground. The fact that the schematic shows the polystyrene cap to be 2500pf (it’s actually a 1000pf) Dannysound has answered.

I show again my way of keeping the matched transistors close together, for a common thermal condition. I put a little heat sink compound between them and used heat shrink wrap to hug them together. The temperature compensating resistor gets mounted over them, in contact.

The above pic also shows the axial lead 1nf polystyrene capacitor (1000J) that I substituted for the one supplied, which has radial leads (see the original build post).

The main PCB, above, is all assembled. It’s also the back of the module with all the trim pots. Yes, I did substitute some capacitors. No worries if you use the supplied ones. Speaking of trim pots, calibration is very easy with this oscillator, especially if you have a scope. I used the Mordax Data in calibration of the 1V/octave trimmer, by using the Voltage Monitor outputs to make +3V and +7V fixed sources, and then switching to the Tuner. What an amazing module, the Data.

Comparing the two oscillators

It’s a great advantage to have two of the same module, so you can compare their behavior. The waveforms of the two are practically identical after trimming. I did find one perplexing difference. The frequency ranges of the oscillators are somewhat different. By comparing the range available with the main frequency knob at four different settings of the Thru-Zero Offset knob, I found that the new EN129 is higher at both the low and high frequency ends. I can’t really explain it, given that the circuits and parts are almost identical. Maybe it’s within the tolerance of the polystyrene integration capacitor. In any case it’s not enough of a difference to worry me.

I mounted my three Dannysound oscillators next to one another, the third being the Cali oscillator, based on the Buchla 258 VCO.