[sdiy] VCO unstable in Spice, stable in real world

[Rutger Vlek]

Hi,

Thanks for all the good suggestions! Decreasing the simulation step time was a good idea. It didn't change a great deal, but made the saw output more precise since the simulation is no longer aliasing on the discharge pulses for the capacitor. I've also replaced the LM311's model and it seems to do its job much better now. I've also found out that the way I prefer Spice is mostly to simulate and understand small parts of a circuit, rather than an entire synth at once :). The VCO I'm trying to model is a single instance of Jurgen Haible's Living VCO. By doing a Spice model I'm hoping to get my electronics knowledge up to speed, such that I can make some modifications. One of the modifications on my list is to add oscillator sync.

I've been wondering by the way. If you take a VCO's rising saw output, invert it and shape it into a pulse that can trigger the second VCO's discharge path, a great deal of delay is introduced. This means that two synced oscillators will actually not reset at exactly the same time. Is this phase-difference considered useful or unwanted? Taking the comparator output of VCO 1 to directly trigger the second VCO's discharge patch would assure much better (phase-) synchronization (although switching sync on and off might me a little more difficult). From the VCO's I know this trick is never used... I wonder why?

> In short, SPICE modeling is well worth the effort, but it can also be very
> tricky, and many of the models don't appear to be very accurate.
> 

Yea, that's just annoying! Also, B2 Spice claims that they have integrated with Eagle PCB. Well... I wouldn't call importing the schematic without working models 'integration'. It seems they have a long way to go, but same holds for the Eagle designers in terms of User Interface by the way :). Ok, that's enough bitching. It's still great that I can play with different components without actually having to buy them (nor blow them...).

Rutger

ps. Speaking of oscillator sync. Has anyone ever tried making an analog VCO that tracks relatively well and correct the remaining tuning error by syncing it to a digital oscillator?