Re: FSUSB Port 0 in OAMP2430 SDP

["Amit Gupta" <amit_jobs05@inbox.com>]

Felipe

Appreciate your prompt reply.

Just to reiterate what you said:-

1. Initiallly on powering on the board, the Vbus voltage is 3.2 volts
(and not 5 volts), this is required to sense the device.
2. Now the configuration needs to be done like enabling the
interrupt(of transceiver and OMAP controller), configuration of OMAP
host controller registers wihin the USB host driver code.
3. Now when device is conntected,  the D+/D- will be pulled up and
hence the transceiver detects some device has beeen connected and
interrrupt is generated by transceiver to the OMAP.
4. Once the interrupt is generated, the transceiver will power up the
5 volts(automatically by reading ID pin) /or the host controller
detecting the interrupt will forcefully power up the Vbus(through
transceiver register) to 5 volts (That we have to find it out - how it
can be done).

Just confirm whether this sequence is correct.

Again thanks in advance.

Thanks & Regards

Ammit

On 8/9/07, Felipe Balbi <felipebalbi@users.sourceforge.net> wrote:
> Hi,
>
> On 8/8/07, Amit Gupta <amitjobs05@gmail.com> wrote:
> > Hi Felipe/Nishan/all
> >
> > I have been able to sucessfully get an USB host interrupt by pulling
> > up voltage levels in Transceiver but its configuring device as low
> > powered device. Also I am getting host interrupt even without
> > connecting the device since pulling up the voltage levels resulted in
> > maybe D+/D- pull up which is sensed as an host interrupt by OMAP
> > board.
> >
> > You seem to be correct in your saying that transceiver does more than
> > just sensing the voltage levels and passig the interrupts. But I am
> > not able to understand the sequence of thing happening like before
> > connecting the device should the port be powered up to 5 volts or it
> > should happen when device connected pulls up/down D+/D- pin. And once
> > this is done then interrupt is also sent and enumeration starts ?
>
> It can only power up 5V after sensing a device connected and its
> pullups are connected. After that, it'll start the enumeration.
>
> Actually, every host/otg controller has to provide means for you to
> force Vbus on, but it'll only work if you have any load between 5V and
> Vbus (try with 300 Ohm resistor) and you can also add a 1K Ohm
> resistor between D+/D-
>
> To understand how to force the controller to power Vbus, read its docs
> since it's completely HW dependent.
>
> >
> > I am going through the USB specs to know this but I suppose one who
> > has worked in USB can immediately answer to my this query since its
> > very urgent.
>
> hopefully I could help you ;-)
>
> >
> > Thanks & Regards
> >
> > Amit
> >
> > ---------- Forwarded message ----------
> > From: Amit Gupta <amit_jobs05@inbox.com>
> > Date: Aug 7, 2007 1:40 AM
> > Subject: Re: FSUSB Port 0 in OAMP2430 SDP
> > To:
> > Cc: Felipe Balbi <felipebalbi@users.sourceforge.net>, Linux OMAP
> > <linux-omap-open-source@linux.omap.com>, "Woodruff, Richard"
> > <r-woodruff2@ti.com>, "Kamat, Nishant" <nskamat@ti.com>
> >
> >
> > >>I think more stuff happens in hardware then just setting those
> > >>registers. Probably the transceiver is applying some pull-up and/or
> > >>pull-down resistors when ID pin is grounded. I'd read musb & otg docs
> > >>again to remember what really happens there.
> >
> > >>If you check OTG specs rev 1.3 it presents what happens when you
> > >>ground ID pin.
> >
> > I was going through OTG specs 1.3. Its difficult to go through
> > completely but then it Pg 23 tells a bit which I was onterested in. It
> > tells that Vbus switcjing can also take place through an application.
> > It says that bus is powered up only when application tries to call ot
> > there is b_srp_init, In the first case if the application is trigering
> > the bus to power on then its definitely happening through software
> > only. Also it talks about bus powered USB device initialting the
> > session. Now it clearly says that it will post an event/interrupt.
> >
> > But still you may be correct that lots of things might be hapening
> > within the transeiver  hardware like pull up/pull down but I  need to
> > know that do that so that I can program the Transceiver accordingly.
> >
> > Thanks in advance
> >
> > Regards
> >
> > Amit
> > On 8/7/07, Woodruff, Richard <r-woodruff2@ti.com> wrote:
> > > > I'm pretty sure you'll need a self-powered hub anyway 'cause MUSB can
> > > > only provide 100mA and if you wanna power a non-otg usb device it uses
> > > > up to 500mA. The self-powered hub masks this limits to the controller
> > > > as it'll power the usb devices itself.
> > >
> > > I think you are both right and wrong.  Its been a few years since I looked, so their probably have been some changes.
> > >
> > > MUSB likely doesn't supply power to anyone.  The board's power supply will do this.  What its capacity is depends on how it was setup.  To be OTG stamp on it a device might need to stay at or below 100mA (might be self powered also).  For the omap-device to get the stamp it might have to at least supply 100.  Do both ends have the USB logo on them?  Do you need them to be?
> > >
> > > An OTG device to get certified need to support many devices ether.  You can have a white list which allows limited support.
> > >
> > > It might be you need to manually turn on some LDO or other on some detection interrupt.  There was one board with the fs-usb block which had this kind of logic I dealt with.  It had 2 ports, one gave just enough power for a target peripheral and another port gave lots of power.
> > >
> > > I seem to recall some noise about raising the limits but that may have been a dream.
> > >
> > > Regards,
> > > Richard W.
> > > _______________________________________________
> > > Linux-omap-open-source mailing list
> > > Linux-omap-open-source@linux.omap.com
> > > http://linux.omap.com/mailman/listinfo/linux-omap-open-source
> > >
> >
>
>
> --
> Best Regards,
>
> Felipe Balbi
> felipebalbi@users.sourceforge.net
>