[sdiy] Driving SSI2144 freq control with PWM from a microcontroller

[Eric Brombaugh]

Correct - range is 1/3 of the original, but much more linear and you can 
usually make up for that with gain elsewhere in the system. I used this 
approach (along with differential outputs) in my FPGA oscillator project:

https://github.com/emeb/up5k_osc

Eric

On 11/5/22 08:28, René Schmitz wrote:
> On 05.11.2022 16:04, Eric Brombaugh via Synth-diy wrote:
>> There's a way to mitigate that - ensure that you always have a high 
>> and low transition in every cycle. Basically you create a very high 
>> frequency PWM signal that's either 33% or 67% depending on the state 
>> of your PDM bit, so the speed of the high & low transitions are always 
>> part of your output and don't vary with the data value. Downside is 
>> that your PDM bitrate has to be 1/3 of that you'd get without doing 
>> this approach.
> 
> If I understand correctly, you also loose 2/3 of the output range. And 
> assumes the t_on and t_off are the same.
>