EmbeddedRelated.com
Forums
The 2024 Embedded Online Conference

Two port implementation for a USB application consuming more than 2.5W

Started by Kieran O' Leary August 11, 2008
Hello,



I am currently working on a USB 2.0 bus powered application that I expect to 
consume more than 2.5W.



As such, I wish to use two USB ports to power the application in a similar 
manner to that used on some USB bus powered external hard drives.



Has anybody had any success in implementing such a scheme and remaining USB 
2.0 compliant? In particular, I wish to "politely" request 500mA from the 
USB host for the second (power-only) port rather than assuming that 500mA 
will be automatically available.



Does anybody know of any circuit implementations available on the web that 
will implement the two port scheme - fully compliant or otherwise?



Any recommendations on how to do it?



Any advice on any gotchas, etc. to watch out for?



Any help would be greatly appreciated.



Many thanks,

Kieran


-- 
Kieran O' Leary
LUTsToDo.com - The website for busy FPGA designers
http://www.lutstodo.com 


>I am currently working on a USB 2.0 bus powered application that I expect to >consume more than 2.5W. >As such, I wish to use two USB ports to power the application in a similar >manner to that used on some USB bus powered external hard drives.
>Has anybody had any success in implementing such a scheme and remaining USB >2.0 compliant? In particular, I wish to "politely" request 500mA from the >USB host for the second (power-only) port rather than assuming that 500mA >will be automatically available.
>Does anybody know of any circuit implementations available on the web that >will implement the two port scheme - fully compliant or otherwise?
>Any recommendations on how to do it?
Ask port1 for 500 mA, and only then ask port2 for another 500mA. And report to the enduser what happends. Like led1 red->green, led2 red->green. If the device doesn't get it's way powerwise let it still be present. But leave powerhungry parts offline.
>Any advice on any gotchas, etc. to watch out for?
IF the potentials of the usb ports doesn't match. Not that I think any usb device does this.. However a user might plug the connectors to two different devices ;) (gotcha!)
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

sky465nm@trline4.org wrote:
> Ask port1 for 500 mA, and only then ask port2 for another 500mA. > And report to the enduser what happends. Like led1 red->green, led2 red->green.
Please don't use red/green indicators for equipment! People with colour blindness will find it impossible to tell the difference. A better way would be to flash the LED when no power is provided and have it continuously on when it is. - -- Brendan Gillatt | GPG Key: 0xBF6A0D94 brendan {a} brendangillatt (dot) co (dot) uk http://www.brendangillatt.co.uk -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (MingW32) iD8DBQFIoZgSuv4tpb9qDZQRAjCNAKCEv0LRBiqlqtooXVap7sjgNx4n3QCgltLl bexETku6RQDCMjR1QGpNolg= =dGhb -----END PGP SIGNATURE-----
"Brendan Gillatt" <brendanREMOVETHIS@brendanREMOVETHISgillatt.co.uk> wrote 
in message news:WcqdnSoFnZqPBTzVnZ2dneKdnZydnZ2d@pipex.net...
> -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > sky465nm@trline4.org wrote: >> Ask port1 for 500 mA, and only then ask port2 for another 500mA. >> And report to the enduser what happends. Like led1 red->green, led2 >> red->green. > > Please don't use red/green indicators for equipment! People with colour > blindness will find it impossible to tell the difference. A better way > would be to flash the LED when no power is provided and have it > continuously on when it is. > > - -- > Brendan Gillatt | GPG Key: 0xBF6A0D94 > brendan {a} brendangillatt (dot) co (dot) uk > http://www.brendangillatt.co.uk > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.7 (MingW32) > > iD8DBQFIoZgSuv4tpb9qDZQRAjCNAKCEv0LRBiqlqtooXVap7sjgNx4n3QCgltLl > bexETku6RQDCMjR1QGpNolg= > =dGhb > -----END PGP SIGNATURE-----
Hi Brendan, This is a reasonable point. As it happens, any indicators that I will use will be software based. Excuse my ignorance but do you know if people with colour blindness have difficulty distingushing colours (red/green) on computer monitors in the same way as with LEDs, etc.? Thanks, Kieran -- Kieran O' Leary LUTsToDo.com - The website for busy FPGA designers http://www.lutstodo.com
<sky465nm@trline4.org> wrote in message news:g7qpdd$n4m$1@aioe.org...
> >I am currently working on a USB 2.0 bus powered application that I expect > >to >>consume more than 2.5W. >>As such, I wish to use two USB ports to power the application in a similar >>manner to that used on some USB bus powered external hard drives. > >>Has anybody had any success in implementing such a scheme and remaining >>USB >>2.0 compliant? In particular, I wish to "politely" request 500mA from the >>USB host for the second (power-only) port rather than assuming that 500mA >>will be automatically available. > >>Does anybody know of any circuit implementations available on the web that >>will implement the two port scheme - fully compliant or otherwise? > >>Any recommendations on how to do it? > > Ask port1 for 500 mA, and only then ask port2 for another 500mA. > And report to the enduser what happends. Like led1 red->green, led2 > red->green. > If the device doesn't get it's way powerwise let it still be present. But > leave > powerhungry parts offline. > >>Any advice on any gotchas, etc. to watch out for? > > IF the potentials of the usb ports doesn't match. Not that I think any usb > device does this.. > However a user might plug the connectors to two different devices ;) > (gotcha!) >
Hi, Thanks ... I think that this is the avenue that I will go down alright although the idea of using a second USB peripheral IC just to enumerate correctly to get the 500mA doesn't seem very cost effective to me. I may try to implement part of the second USB interface in some spare capacity I may have available on an FPGA. A possible potential difference between the two 5V voltage sources is something that will require some thought alright but I have some ideas for this. Thanks again. Cheers, Kieran -- Kieran O' Leary LUTsToDo.com - The website for busy FPGA designers http://www.lutstodo.com
Kieran O' Leary wrote:
> "Brendan Gillatt" <brendanREMOVETHIS@brendanREMOVETHISgillatt.co.uk> wrote >
... snip ...
> >> Please don't use red/green indicators for equipment! People with >> colour blindness will find it impossible to tell the difference. >> A better way would be to flash the LED when no power is provided >> and have it continuously on when it is. > > This is a reasonable point. As it happens, any indicators that I > will use will be software based. > > Excuse my ignorance but do you know if people with colour > blindness have difficulty distingushing colours (red/green) on > computer monitors in the same way as with LEDs, etc.?
There are all sorts of colour blindness'. I have one, and rarely notice it. Some weird text colour schemes, including green type on some sort of background, are very hard for me to see. -- [mail]: Chuck F (cbfalconer at maineline dot net) [page]: <http://cbfalconer.home.att.net> Try the download section.
Kieran O' Leary wrote:
> > I am currently working on a USB 2.0 bus powered application that > I expect to consume more than 2.5W. As such, I wish to use two > USB ports to power the application in a similar manner to that > used on some USB bus powered external hard drives. >
... snip ...
> > Any recommendations on how to do it?
Just ensure that you load the two supplies separately. I.e. don't parallel the two supplies. Since their output voltages will not be equal, the result would be that one supplied all the current. -- [mail]: Chuck F (cbfalconer at maineline dot net) [page]: <http://cbfalconer.home.att.net> Try the download section.
On 2008-08-12, Kieran O' Leary <kieran@lutstodo.com> wrote:

> Excuse my ignorance but do you know if people with colour > blindness have difficulty distingushing colours (red/green) on > computer monitors in the same way as with LEDs, etc.?
Yes. Somebody who is red-green color-blind (the most common form of color-blindness) will have pretty much the same problems with a CRT and with LEDs. -- Grant Edwards grante Yow! Okay... I'm going at home to write the "I HATE visi.com RUBIK's CUBE HANDBOOK FOR DEAD CAT LOVERS"...
In article <3uSdnTBaTdD91z_VnZ2dnUVZ_uidnZ2d@posted.usinternet>, 
grante@visi.com says...
> On 2008-08-12, Kieran O' Leary <kieran@lutstodo.com> wrote: > > > Excuse my ignorance but do you know if people with colour > > blindness have difficulty distingushing colours (red/green) on > > computer monitors in the same way as with LEDs, etc.? > > Yes. Somebody who is red-green color-blind (the most common > form of color-blindness) will have pretty much the same > problems with a CRT and with LEDs.
Red/green colour blindness most common and most often forgotten I once had a link to web page about various colour pairs that should be avoid for most things especially web pages to avoid these issues. Another word of caution that everybody might like to remember in some application fields of which medical is one ONLY use RED indicators to signify events like - Danger (that does NOT mean power on) - Fault with the equipment - Patient EMERGENCY (e.g. heart stopped) It seems a shame that the easiest and cheapest LED colour for manufacturers is the one you should use least. -- Paul Carpenter | paul@pcserviceselectronics.co.uk <http://www.pcserviceselectronics.co.uk/> PC Services <http://www.pcserviceselectronics.co.uk/fonts/> Timing Diagram Font <http://www.gnuh8.org.uk/> GNU H8 - compiler & Renesas H8/H8S/H8 Tiny <http://www.badweb.org.uk/> For those web sites you hate

The 2024 Embedded Online Conference