[sdiy] Electrosmith Eurorack DSP platform

Eric Brombaugh ebrombaugh at gmail.com
Thu Jan 27 22:57:19 CET 2022


I was seeing harmonics down slightly better than 80dB with all the 
mitigation measures active. Without it was more like 60-65dB. Not quite 
as good as even an inexpensive sigma-delta I2S audio DAC, but a few SMT 
passive parts beat that for cost & simplicity and I2S DACs don't support 
12MHz sample rates.

Eric

On 1/27/22 14:44, Mike Bryant wrote:
> Thanks - that makes obvious sense now you mention it !!    P channel MOSFETs turn on very differently to N-channel - your method cancels this out.
> 
> Can I ask what THD did you get to ?
> 
> 
> 
> -----Original Message-----
> From: Synth-diy [mailto:synth-diy-bounces at synth-diy.org] On Behalf Of Eric Brombaugh via Synth-diy
> Sent: 27 January 2022 21:30
> To: synth-diy at synth-diy.org
> Subject: Re: [sdiy] Electrosmith Eurorack DSP platform
> 
> No secrets - this was a personal project to see how well it works.
> Details are here:
> 
> https://github.com/emeb/up5k_osc
> 
> TL;DR on the differential PDM: The generic FPGA output drivers have a bit of asymmetry in the rise/fall times which can cause some low level harmonic distortion. There are a couple of techniques for dealing with this - 1) ensuring there's always a rise and fall in every output period, and 2) using differential signalling to cancel the asymmetry. I tried various combinations and saw improvements for both. Overall I'd estimate there was about a 15-20dB improvement in THD.
> 
> The output filter consists of a passive RC on each differential leg, then feeding into a diff-amp with additional 1-pole filtering that outputs a unipolar signal at modular synth levels. Since the clock rate was so high and the baseband signal bandwidth fairly low in comparison there wasn't much need for fancier filtering than that.
> 
> Eric
> 
> On 1/27/22 13:51, Mike Bryant wrote:
>> Ok Eric, you've got me on this one ! :-)  If it's not a trade secret, what does the output filter look like such that differential PDM is needed, and how much improvement does it give over single ended ?




More information about the Synth-diy mailing list