Pulse Wave Modulation Question

[thewaag]

I was playing around with my LP this weekend, and one of the patches that I was using for a starting point utilized Pulse Wave Modulation. If I remember correctly, the name of the patch was "PWM 2 Pole" or something like that.

It has a very rich Pulse Wave Modulation, kind of what I used to do on my old ARP Odyssey to fatten up the string sounds.

This patch had a fatter PWM sound than I remember from when I first got my LP and started playing with sounds.

As it turns out, this patch is using the filters in 2-pole mode--when you change it to 4-pole mode, the PWM sound is completely lost. I had my LP filters set up for 4-pole originally, so this explains why I did not have that fattness when I first started playing my LP.

Why does 2-pole PWM sound so much fatter than 4-pole PWM??

[LWG]

Hello,

Changing the pole taps on a filter should not affect the density of the instruments overall tone. It does alter the texture of the patch you have pulled up.
The lower number of poles you use (earlier the filter tapped), the greater the percentage of harmonics outside of the cutoff frequency that are allowed to leak thru the filter.
Soundwise, this just means the filter should sound brighter and more buzzy when you select the 2-pole (-12db/oct) mode.
If your patch sounds thicker, you may want to check to see if you have changed another parameter such as pre-filter gain, etc..


Regards,


Lawrence

[electrical_engineer_gEEk]

Just for future reference PWM = Pulse Width Modulation
Article:
http://en.wikipedia.org/wiki/Pulse-width_modulation

[thewaag]

Just for future reference PWM = Pulse Width Modulation
Article:
http://en.wikipedia.org/wiki/Pulse-width_modulation

Yeah, I knew that.....I don't know where the Pulse WAVE Modulation came from. Just trying to rush out the question without really thinking, I guess.

Still hear a big difference between 2-pole and 4-pole without changing any other parameters.....I tried it again last night.

[robles]

The real question is how big a difference there is when your cut-off frequency is all the up (clockwise). When you're not really doing any filtering except up into the inaudible range, there shouldn't be a big difference in the tone. If there is, something wierd is going on.

[Tangsonghe]

Hi,

I'm not much of a filter expert but when you switch between
two and four poles the phase shift will be different (regardless
of cutoff frequency). I can't think of any components downstream
that will care unless you mix with the original signal but here is a madcap theory:

If you break a PWM wave down into its fourier components, each component will have a phase sweep when you vary the duty cycle. You have two PWM waves mixed into the filter. I can envisage a couple of scenarios where the filter phase shift acts differently on the two input PWMs. This will happen if the phase shift is non linear because of factors such as slew rate limits etc. The Moog ladder filter is in engineering terms far from ideal as a filter so to me its possible that it will give a different result depending on the phase shift characteristics of 2 Pole vs 4 Pole, hence the different sound.

Ah, I love coming up with untested theories!