XS1-U8A Power Supply

Technical discussions around xCORE processors (e.g. xcore-200 & xcore.ai).
Post Reply
frido1
Member
Posts: 9
Joined: Fri Jan 08, 2016 12:42 pm

XS1-U8A Power Supply

Post by frido1 »

Hi all,
For a new design I would like to supply VDDCORE and VDD1V8 of the the XS1-U8A by external linear regulators instead of using the internal switching regulators.

1) Are there any numbers for the current consumption of VDDCORE and VDD1V8 when running a typical audio application?

2) Is VSUP only used to supply the internal switching regulators? So, if VDDCORE and VDD1V8 are provided externally, may VSUP be left open / connected to GND?

Many thanks in advance.


henk
Respected Member
Posts: 347
Joined: Wed Jan 27, 2016 5:21 pm

Post by henk »

Hi Frido,

Just a word of warning - you are going off-piste.

Refer to any of the XS1-L datasheets for the 1V supply - expect in the order of 200-250 mA @ 1V for USB audio, depending on environmental conditions, sample rates, amount of DSP, and other things you are doing. 1V8 consumes much less - something like 15-50 mA I believe. It powers part of the USB PHY and the Oscillator.

You probably want to measure both.

I would avoid a linear regulator from 5V to 1V - it would have to get rid of 1 W.

Is there a reason to not provide power to VSUP?

Cheers,
Henk
frido1
Member
Posts: 9
Joined: Fri Jan 08, 2016 12:42 pm

Post by frido1 »

Hi Henk,
Thanks for your answer. Power dissipation is not an issue for this design. I just want to avoid the switchers for EMI reasons (call me weird... ).

So, can anyone confirm that VSUP is exclusively used by the internal switching regulators and consequently may be left open / connected to GND if they are not used?
henk
Respected Member
Posts: 347
Joined: Wed Jan 27, 2016 5:21 pm

Post by henk »

Hi Frido

Even if you supply VSUP, the switching-supplies ought to not switch; if the 1V is slightly too high, the FET should be permanently off, if it is too low, it should be permanently low. Make your external supply 1.1V then the FET ought to be permanently off.

Similar for 1.8V.

It is easily tried with a startKIT and a bench supply - remove the inductor in the supply, and connect a 1.0V bench supply to the chip instead; you will see that the other pad of the inductor is permanently low or high depending on the precise level of the bench supply.

One other word of warning: be aware that you will need some power supply sequencing; this is normally taken care off by the on-board switchers. 3V3 should be on (which is switched internally), before 1V comes up.

Cheers,
Henk
Post Reply