It’s a long way from anything testable, but did want to mention I’ve started work on a phase distortion/pulsaret hybrid oscillator, that should hopefully give some new expressive timbral seasonings to experiment with; if anybody’s interested in helping me run it through the wringer when the time comes, or just intrigued, feel free to reply. The planned release will be a free module, so no delusions of grandeur; if it’s successful, and a few other ideas on the back burner also succeed, then maybe the idea of support beyond caveat emptor won’t be so daunting 
there’s your module name. 
I demand payment in new old stock fluorescent shag toilet seat covers, nothing manufactured after 1976, no refunds!
I assume it will be implemented as a Pulsar synthesis oscillator where each period/cycle is the concatenation of a pulsarette followed by silence in some (controllable) ratio (which Curtis Roads called ‘pulsaret-width modulation’) where the pulsarette is constructed/synthesized from a (controllable) phase distorted source signal/wave shape (sort of Casio PD style synthesis)?
Hi, thanks for asking! It’s actually going to be more the phase distortion “knee” warping will be applied to the pulsaret burst window- there’s a Macro and Micro (N times warp per window) mode division; polyBLAMP will solve some aliasing-vs-CPU considerations, Gibbs ringing/overshoot. There’s five floats to be concerned with: f0, dutyCycle, formantMult, pdKnee, and sampleRate.
Does this help you picture it?
That’s pretty much in line with what I described as ‘where the pulsarette is constructed/synthesized from a (controllable) phase distorted source signal/wave shape’.
There are many ways to control/modulate the phase distortion of the pulsarette and the relative ratio for ‘pulsarette’ and ‘silence’ (for Pulsar synthsesis). Phase distortion and/or pulsar synthesis are relatively uncommon but interesting techniques.
It was Hamburg Audio’s Nuklear synth that first intrigued me re: pulsaret synthesis; often more exotic types of synthesis have a limited timbral flexibility- I thought Scanned Synth/Emzyme was such a cool concept, but so much inharmonicity left the use cases kinda “Do you want to remind people of a giant, rusted box-spring mattress being slapped and bowed?” Verklagekasper’s Discrete Summing Formulae-based synths are great implementations, but nobody’s likely to be using Sonitarium to score children’s programming… all that to say, both PD and pulsaret synthesis have a broader palette, so the result shouldn’t be a sonic white elephant