aaa Square PWM - Synths & synthesis forums
skin: 1 2 3 4 |  Login | Join Dancetech |

dancetech forums

18-May-2024

Info-line:   [synths]    [sampler]    [drumbox]    [effects]    [mixers]     [mics]     [monitors]    [pc-h/ware]    [pc-s/ware]    [plugins]    -    [links]    [tips]

Search forums House rules Live chat Login to access your admin About dancetech forums Forum home Start a new topic

Forums   -   Synths & synthesis

Subject: Square PWM


Viewing all 11 messages  -  View by pages of 10:  1 2


Original Message 1/11             11-Apr-00  @  07:57 AM   -   Square PWM

frankreitz

Posts:

Link?: Link

File?:  No file



I don't have a tb303 myself but I've been studying some wave files of this phenomenon.
It seems that the square wave of the tb is pulse width modulated. On some waves more than other. Does anyone out there know about this?
Thanks,
Frank..



[ back to forum ]              [quote]

Message 2/11             11-Apr-00  @  03:04 PM   -   RE: Square PWM

R

Posts: 9

Link?:  Link

File?:  No file



Hhmm, this is possible cause the 303 doesn't own two independant oscillators circuits for saw and square, but it uses a simple transistor amplifier stage to convert the saw into square. A simulation confirms that the pulse width is getting a bit smaller at higher frequencies. Correct me if i'm wrong.



[ back to forum ]              [quote]

Message 3/11             12-Apr-00  @  01:28 AM   -   RE: Square PWM

frankreitz

Posts:

Link?:  No link

File?:  No file



In reply to R@F. What you suggest as a possible cause could very well be true. But would this transistor amp create the pwm because it's more or less faulty (I've seen Heavy PWM on some waves, like 20/80%) or is there something else that can affect this transistor in order to get this heavy pwm.
As to the pwm getting smaller at higher frequencies holds pretty much true against what i've seen if you refer to cutoff as frequency! Correct _me_ if I'm wrong...
Frank



[ back to forum ]              [quote]

Message 4/11             12-Apr-00  @  01:28 AM   -   RE: Square PWM

frankreitz

Posts:

Link?:  No link

File?:  No file



In reply to R@F. What you suggest as a possible cause could very well be true. But would this transistor amp create the pwm because it's more or less faulty (I've seen Heavy PWM on some waves, like 20/80%) or is there something else that can affect this transistor in order to get this heavy pwm.
As to the pwm getting smaller at higher frequencies holds pretty much true against what i've seen if you refer to cutoff as frequency! Correct _me_ if I'm wrong...
Frank



[ back to forum ]              [quote]

Message 5/11             12-Apr-00  @  12:05 PM   -   RE: Square PWM

R

Posts: 9

Link?:  Link

File?:  No file



Just plugged the 303 to my scope. First thing: The waveform on the output is far away from a perfect square. It looks like being put through a highpass filter. But it is in fact highpass filtered cause of the very small decoupling capacitors (0.01µF) between VCF and VCA. Everyone knows the mod to increase C21 , C22 for better bass response. So it's a bit hard to see any PWM on the screen. I've simulated the saw2square converter with EWB and found out that the pulse width is decreasing at higher oscillator frequencies (not cutoff). It's about perfect 50% at 100Hz and drifts towards 40% at about 1000Hz. I don't think, this effect is noticeable by ear cause there are more than 3 octaves between these frequencies. So what was the reason for the PWM on the screen?? You said that the PWM is changing when tweaking Cutoff. Be sure that Resonance is set to 0 when checking the waveforms! This is important cause you will see the cutoff frequency as a sine oscillation in the waveform. The level of this oscillation depends on the Resonance setting. Maybe this gave the effect of a PWM.



[ back to forum ]              [quote]

Message 6/11             13-Apr-00  @  01:05 AM   -   RE: Square PWM

frankreitz

Posts:

Link?:  No link

File?:  No file



I know about the resonance/clean waveforms and how they look on a graph. In fact, the high frequent resonance does not disturb the picture, where, on the other hand the lower frequent resonance does and makes it almost impossible to determine the pwm. I've measured pwm on no resonant waves and high frequency resonance waves!
I've got a square-wave example of a _very_ low frequent sound (without resonance ). It has a mark space ratio at about 20/80% But on the other hand, it's distorting a bit (not a pedal, just move the volume knob fully clockwise and it'll distort!), so maybe that, too, could affect the saw2sqr curcuit?
How exactly does the Saw2Sqr curcuit work? And what about the frequencies below 100Hz, do they then go towards 60%? I know 100Hz is low but just out of couriousity.
The reason for all this is that I'm trying to program a 303-emulator. I want to include as many things from the original as possible and as a fact the pwm really is audible!



[ back to forum ]              [quote]

Message 7/11             16-Apr-00  @  06:21 PM   -   RE: Square PWM

R

Posts: 9

Link?:  Link

File?:  No file



The square2saw convertor is just a simple common emitter amplifier. It has a gain of about 30dB which overdrives the saw waveform and produces a square. The pulse width is just slightly changing below 100Hz. It reaches something about 55% at 50 Hz if the simulation is right.
The 303 is always distorting a bit when tweaking the volume knob full clockwise. Especially when using the headphone output . But I belive this is not noticable on the main output.
Nevertheless I wasn't able to see such a heavy PWM of about 20% on my scope. Maybe it depends on the sequence you're using or something else. I don't know! The only thing I can definetly say is that the oscillator for itself shouldn't be theoretically able to produce such a heavy PWM. Can you hook me up with some waveforms or something else?



[ back to forum ]              [quote]

Message 8/11             18-Apr-00  @  07:59 AM   -   RE: Square PWM

frankreitz

Posts:

Link?:  No link

File?:  No file



Tried to mail you a wave-file, but my mail hot bounced back. What address should I send it to?



[ back to forum ]              [quote]

Message 9/11             19-Apr-00  @  04:46 PM   -   RE: Square PWM

R

Posts: 9

Link?:  Link

File?:  No file



Try this:
network-909@uni-duisburg.de

thanks in advance



[ back to forum ]              [quote]

Message 10/11             19-Apr-00  @  09:09 PM   -   RE: Square PWM

ReZo-99

Posts:

Link?:  No link

File?:  No file



On my 303, my square wave is a bit more on the saw side. I mean, its not as smooth as I want it. How can I adjust it to get a smoother square wave? I tried adjusting the PWM inside but I still don't have the sound I want.

ReZo-99



[ back to forum ]              [quote]

Message 11/11             24-Apr-00  @  01:35 PM   -   RE: Square PWM

R

Posts: 9

Link?:  Link

File?:  No file



This is mainly due to the small decoupling capacitors between VCF and VCA. Just increase C21,22 and the waveform should look more like a square You'll also get a better overall bass response.



[ back to forum ]              [quote]

Viewing all 11 messages  -  View by pages of 10:  1 2

There are 11 total messages for this topic





Reply to Thread

You need to register/login to use the forum.

Click here  to Signup or Login !

[you'll be brought right back to this point after signing up]



Back to Forum





Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko; compatible; ClaudeBot/1.0; +claudebot@anthropic.com)