[sdiy] Hypertriangularity and the 3340

[Tom Wiltshire]

> On 6 Oct 2017, at 17:40, KA4HJH <ka4hjh at gmail.com> wrote:
> 
> 
>> On Oct 6, 2017, at 11:23 AM, Tom Wiltshire <tom at electricdruid.net> wrote:
>> 
>> Alternatively, in these modern days of plenty, you could use a 3340!!
> 
> As Craig Anderton did with the HyperFlanger (or was one of Thomas Henry's?). IIRC, the main reason he did this was to be able to feed the sawtooth output back into the VCO to produce an exponential sawtooth wave. This eliminated the need for expo conversion on the input to the 4046 to sweep the comb filter response exponentially instead of linearly. I've never heard this effect but it *sounds* interesting. 
> 
> Um, I *think* he used a 4046 for the clock. I can't remember now but the trick would work with any linear HF-VCO circuit.
> 
> My HyperFlanger manual is lost in a pile somewhere around here, along with most of my thoughts. 8/
> 
> 
> Terry Bowman, KA4HJH
> "The Mac Doctor”

I implemented an exponential clock modulation on the PIC-based flanger clock chip I did. It’s not a *hugely* noticeable effect, but it makes the LFO waveforms sound more like you expect them. Without it, they tend to linger at the top, sweeping through the endless Hertz of upper octaves before briefly sweeping the limited Hertz of lower octaves. That distorts the wave shape, at least according to human hearing. Of course, on a PIC, I could equally well have implemented an LFO with a hyper-triangle wave (I did exactly that once before) but in this instance, I wanted to allow the possibility of external modulation with the exponential response.

http://electricdruid.net/flangelicious-a-super-dooper-flanger/

One PIC is a lot simpler than a 3340, a 4046, and a 4041, although I could only get a 40:1 sweep, whereas Anderton’s more sophisticated solution manages 60:1, which is definitely at the top end of flangers.

Tom