Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Steve-White]

Yes, exactly right. That was my (kinda joking) point. DSD is PWM. My Powersofts switch at 460 kHz which, I believe, puts them in an ISM band and cuts them some slack with regard to Part 15  –  radio folks can confirm. Going straight from DSD to speakers puts the switching frequency up in the MHz. I was just thinking in terms of audiophools and something stupid for them to get excited about. You can take all this with a grain of salt as we just knocked down a couple of Martinis in honor of Olivia Newton-John.

--Frank

PS: Delta-sigma mod, correctly called sigma delta-modulation (I looked it up), puts an integrator in front of the delta-modulator. But what’s an integrator between friends?

RIP to ONJ

[Henry Cohen]

Yes, exactly right. That was my (kinda joking) point. DSD is PWM. My Powersofts switch at 460 kHz which, I believe, puts them in an ISM band and cuts them some slack with regard to Part 15  –  radio folks can confirm.

460kHz has no Part 18 (ISM) allocation. It does however fall into:
- Part 80, maritime band. International aeronautical radio navigation & radio telegraphy.
- Part 15, generally power line carrier (PLC) communications used be electric utilities.

[Brian Jojade]

By definition "each point in time" is sampling, so what you're suggesting is that we sample slope instead of amplitude. In that case, if I'm remembering my high school calculus correctly, "measuring" the slope of a waveform at a given point in time is way more difficult (maybe technically impossible? My calculus is fuzzy) than simply measuring amplitude.

A point doesn't have a slope. It's just a point.  In order to determine a slope, you need 2 points so that you have a starting point and an ending point.  Now that slope can either be a straight line, or it can be a curved line.  For something simple like a sinewave, you could have very few points and recreate a curve with a relatively simple formula.  For musical content, to be able to recreate a curve that matched the signal would be astronomically difficult at best, and would require sampling a period of music in order to create that calculation.  It couldn't be done in real time like can be done with sampling many points.

[Frank Koenig]

460kHz has no Part 18 (ISM) allocation. It does however fall into:
- Part 80, maritime band. International aeronautical radio navigation & radio telegraphy.
- Part 15, generally power line carrier (PLC) communications used be electric utilities.

Thanks Henry. My mistaken belief came from designing electrosurgical generators back in the ‘90s. We used 450 or 460 kHz as I recall and I thought we had some special dispensation there. It was likely the prohibited band from 490-510 kHz that we were staying clear of (18.303). Our devices were clearly ISM but we were operating outside ISM bands under 18.301. As I read it now we would have been allowed 25 uV/m at 300 m radiated emissions and there’s really nothing in Part 15 that would have applied except maybe to unintentional emissions from the computer part of the system. I kind of love and hate this stuff -- and in my retirement enjoy that making noise with speakers is largely unregulated.

--Frank

[Russell Ault]

A point doesn't have a slope. It's just a point. {...}

Differential calculus would beg to differ (but, as I say, good luck measuring it).

-Russ

[ProSoundWeb Community]