The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> High-Power Design >> "low-interference" DCDC-converter
https://designers-guide.org/forum/YaBB.pl?num=1207830578

Message started by Berti on Apr 10^th, 2008, 5:29am

Title: "low-interference" DCDC-converter
Post by Berti on Apr 10^th, 2008, 5:29am

Hi all,

I am working on a DCDC-converter for a SOC. It is important that the converter generates
low interference (e.g. switching noise). But most of the publications about DCDC-converter focus on
efficiency (which is secondary in my case, but a LDO is still not good enough).
Can somebody give me suggestions (literature) about "low-interference" DCDC-converters (e.g. using dithering etc.)?

Thanks

Title: Re: "low-interference" DCDC-converter
Post by sheldon on Apr 11^th, 2008, 4:43pm

Berti,

Some designers, at least at the university level, are starting to use Sigma-Delta
controllers for DC-to-DC Converters to reduce noise and improve performance. Still
seems a little experimental but it is an interesting approach.

http://users.ece.gatech.edu/rincon-mora/publicat/journals/iecon05_tran_bypss_bst.pdf

Best Regards,

Sheldon

Title: Re: "low-interference" DCDC-converter
Post by loose-electron on Apr 15^th, 2008, 12:39pm

You might want to post some more specifics about what you want to do. Any switching converter is going to be noisier than a linear regulaotr (classic, LDO or whatever) but need to know more about your system architecture requirements.

Title: Re: "low-interference" DCDC-converter
Post by Berti on Apr 15^th, 2008, 10:44pm

Thank you Sheldon for the reference. I found some interesting literature.

Loose-electron,

The dcdc-converter is for a wireless transceiver. As the circuits operate from a lower supply than the battery voltage, a dcdc-conveter is mandatory for low power consumption (and
the efficiency of a LDO is unfortunatly not acceptable).
First measurements have already been quite good. But the trasmitter is very sensitive and the interference from the dcdc-converter should be reduced further by 20dB.
I think it is difficult to give absolute specs for isolation since the some of the coupling paths have been identified, but they are difficult to model.

So I was wondering what other techniques I can incoorporate into the existing dcdc-converter (PWM buck converter) to further reduce the switching noise.

Cheers

Title: Re: "low-interference" DCDC-converter
Post by sheldon on Apr 16^th, 2008, 6:26am

Berti,

This is probably not the direction you are looking for and is more of a philosophical
question. However, it if the design is so sensitive to noise, does integration make
sense? I don't understand the economics driving your project, but would System
in Package, SiP, be a better approach than integration for this level of performance?
SiP would give you an opportunity to use passives to reduce noise, etc. that is
difficult to do in integrated designs.

Best Regards,

Sheldon

Title: Re: "low-interference" DCDC-converter
Post by Berti on Apr 16^th, 2008, 10:30pm

Sheldon,

SiP and solutions with external power switches have also been considered.
But BOM and customer demands don't allow that.

Regards

Title: Re: "low-interference" DCDC-converter
Post by sheldon on Apr 25^th, 2008, 6:38pm

Berti,

My office is moving and while cleaning up my desk found another paper
that may be useful.

"All-Digital DPWM/DPFM Controller for Low-Power DC-DC Converters"
by Wang, Rahman, Lukic, Prodic

It describes an all-digital dc-to-dc converter. Since the dc-to-dc converter
operates at >6MHz. Operating at that frequency should keep the ripple at
the output down and out of the audio band. There is also a pretty extensive
bibliography you can use to study the prior art.

Best Regards,

Sheldon