Forums

power on firewire port ?

Started by Robert Lacoste January 13, 2004
There is power available on the 6-pin firewire connectors, but how much ?
Does the standard gives a precise guaranteed minimum available power (shared
by the slaves of course) ?

Thanks for your help,

Robert Lacoste - ALCIOM
http://www.alciom.com



Robert Lacoste wrote:
> There is power available on the 6-pin firewire connectors, but how much ? > Does the standard gives a precise guaranteed minimum available power (shared > by the slaves of course) ?
Section 4.2.2.7 of the 1394 spec details the power availability. In 1394, a node could provide a maximum current of 1.5A with a maximum voltage of 40V. The self-id packets give detail about what a node provides and requires from a bus in terms of power. The 1394a spec has changed some of these values (e.g. max voltage is now 30V). There is no guarantee, however, that there is a node on the bus that provides power. -- Paul Black mailto:paul.black@oxsemi.com Oxford Semiconductor Ltd http://www.oxsemi.com 25 Milton Park, Abingdon, Tel: +44 (0) 1235 824 909 Oxfordshire. OX14 4SH Fax: +44 (0) 1235 821 141
MANY THANKS !

That was the missing point for me : there isn't any guarantee that a node
will provide power.
So for example (in my case...) the developper of a Firewire peripheral, to
be connected to a PC, can't rely on a power source from the PC at all, right
?

Robert

"Paul Black" <paul.black@nospam.oxsemi.com> a &#2013265929;crit dans le message de
news:bu0pjl$6rb$1@jupiter.oxsemi.com...
> Robert Lacoste wrote: > > There is power available on the 6-pin firewire connectors, but how much
?
> > Does the standard gives a precise guaranteed minimum available power
(shared
> > by the slaves of course) ? > > Section 4.2.2.7 of the 1394 spec details the power availability. In > 1394, a node could provide a maximum current of 1.5A with a maximum > voltage of 40V. The self-id packets give detail about what a node > provides and requires from a bus in terms of power. The 1394a spec has > changed some of these values (e.g. max voltage is now 30V). > > There is no guarantee, however, that there is a node on the bus that > provides power. > > -- > Paul Black mailto:paul.black@oxsemi.com > Oxford Semiconductor Ltd http://www.oxsemi.com > 25 Milton Park, Abingdon, Tel: +44 (0) 1235 824 909 > Oxfordshire. OX14 4SH Fax: +44 (0) 1235 821 141
Robert Lacoste wrote:
> MANY THANKS ! > > That was the missing point for me : there isn't any guarantee that a > node will provide power. > So for example (in my case...) the developper of a Firewire > peripheral, to be connected to a PC, can't rely on a power source > from the PC at all, right ? >
Does it matter though? If you're unpowered, you will not get any bus events or messages; if you do get bus events or messages then you are powered. Some laptops only have a 4-pin 1394 port; these do not supply power. I'm not aware of any PC or Mac that does not supply power. For all intents and purpose, there will always be at least one power-provider capable node on a bus. Furthermore, you are allowed (assured of) up to 3W for the self-id phase where you should report how much power you actually require. There are three 1394 Trade Association specifications in connection with 1394 power; you can order these on-line at http://www.1394ta.com/ and receive a PDF the next day: 1394 TRADE ASSOCIATION Power Specification Part 1: Cable Power Distribution Doc1999001_Part1.pdf 1394 TRADE ASSOCIATION Power Specification Part 2: Suspend/Resume Implementation Guidelines Doc1999001_Part2.pdf TA Document 1999001-3 Power Specification Part 3: Power Distribution Management Doc1999001_Part3.pdf
In article <40041ecb$0$237$4d4eb98e@news.dk.uu.net>, 
nospamtcl@nospamphaseone.nospamdk says...
> Some laptops only have a 4-pin 1394 port; these do not supply power.
Correct.
> I'm not aware of any PC or Mac that does not supply power.
Apart from the aforementioned laptops.
> For all intents and purpose, there will always be at least one power- > provider capable node on a bus.
This is not true. There is nothing to prevent you from attaching a device to a notebook that expects power and not find it there. Ask me how I know. (It sickens me that nobody makes a notebook with a 6-pin powered 1394 port anymore, it makes those ultra-small "drop in your shirt pocket" external drive carriers useful for notebooks in the field. Grrr. It's not exactly easy to plug in a wall-wart when you're in the middle of nowhere, with a notebook with plenty of battery power, but no 6-pin firewire. However, you can get PCMCIA 1394 cards with 6-pin ports, but they really suck the power, and chew up an extra card slot. -- Randy Howard 2reply remove FOOBAR
Randy Howard <randy.howard@foomegapathdslbar.net> wrote:

> the field. Grrr. It's not exactly easy to plug in a wall-wart when > you're in the middle of nowhere, with a notebook with plenty of > battery power, but no 6-pin firewire. However, you can get PCMCIA > 1394 cards with 6-pin ports, but they really suck the power,
Well, that power-sucking aspect is quite possibly the reason why so many laptops refuse to power Firewire devices --- it just consumes too much of that most precious commodity in a laptop: battery power. In case of dire need, you could always use a USB external instead --- that's defined to always be powered by the host. If you use USB-2.0, you'll even get the same kind of speed as with 1394. Or, if you really want to stick 1394, there's bound to be some shop that makes a slave-power interface that connects to, say, both USB and 1394 on the laptop side and uses those to provide powered 1394. -- Hans-Bernhard Broeker (broeker@physik.rwth-aachen.de) Even if all the snow were burnt, ashes would remain.
Hans-Bernhard Broeker wrote:
> Randy Howard <randy.howard@foomegapathdslbar.net> wrote: > >> the field. Grrr. It's not exactly easy to plug in a wall-wart when >> you're in the middle of nowhere, with a notebook with plenty of >> battery power, but no 6-pin firewire. However, you can get PCMCIA >> 1394 cards with 6-pin ports, but they really suck the power, > > Well, that power-sucking aspect is quite possibly the reason why so > many laptops refuse to power Firewire devices --- it just consumes too > much of that most precious commodity in a laptop: battery power.
There's nothing power hungry about the FireWire bus itself; a powered, inactive bus consumes only mW. The problem is that some FireWire devices can require lots of power (tens of Watts); this would strain the laptop somewhat. Some manufacturers seem to have taken the easy option of not supplying any power. Personally, I think this is a bit of a crude, inconsiderate solution; most laptops could easy provide power for the type of device that one would connect to a laptop... and if a particular device requires to much power, then the laptop could always refuse to power it (either by not turing the device's link on or by current limiting).
> In case of dire need, you could always use a USB external instead --- > that's defined to always be powered by the host. If you use USB-2.0, > you'll even get the same kind of speed as with 1394. > > Or, if you really want to stick 1394, there's bound to be some shop > that makes a slave-power interface that connects to, say, both USB and > 1394 on the laptop side and uses those to provide powered 1394.
> battery power, but no 6-pin firewire. However, you can get PCMCIA > 1394 cards with 6-pin ports, but they really suck the power, and > chew up an extra card slot.
And PCMCIA FireWire cards often don't provide bus power anyway, as I found out to my detriment when I first acquired my iPod. Plus, all the PCMCIA FW cards I've seen have a dirty great extension on the end holding the FireWire connectors, so they often block your other card slot anyway (no XJack devices can coexist with them, for instance). Laptop vendors seem to be targeting direct connect to DV camcorders as the "killer app" for 1394 on PCs, hence the miniature and unpowered connector. FireWire still has a "taste" of Macintosh to PC vendors, I think. USB (and USB 2.0, now) are the "PC way".