Netbooks with DVD
in [UK Linux]
|
From: Johnny B Good on 12 Dec 2008 20:01 The message <VA.000015e8.820450f9(a)nospam.aaisp.org> from Daniel James <wastebasket(a)nospam.aaisp.org> contains these words: > In article news:<slrngk28mc.jhj.news06(a)desktop.tarcus.org.uk>, Ian > Rawlings wrote: > > > I have an LG "USB Powered" DVD writer. > > ... > > I'll bear that in mind personally as I find my Plextor's requirement > > for two USB ports to be a pain, however it is a very good drive in > > all other respects, dealing with things that other drives turn their > > noses up at. > Yup. Plextor certainly do have a good reputation. > > > Another thing to watch out for is that the USB ports of a > > > laptop/netwook may share their power input ... > > ... > > I've never come across an *in spec* situation like that ... > Did I say it was in spec? Fact is: there are PCs that are built like > that. > > ... you might find that a port will supply 1A but if you use two, > > each port will only supply 500mA ... > Or, perhaps nearer the truth, the two ports can supply 1A between them, > to whatever devices may be attached. I'm sure I've seen boards on which > two USB ports were protected by a single overload 'fuse'. > > I'd imagine that's what you're alluding to, and your LG drive > > will draw more than 500mA from a port if it's available. > It may do ... wouldn't surprise me when writing. > http://uk.lge.com/products/model/detail/dvdrewriter_gsae50n.jhtml# > doesn't say ... not does the (brief) pdf manual. All it says is: > � The Rewriter may get its power from the USB power. > All systems may not meet USB power requirements and > full performance may not be achieved. In this case, > connect to the included USB power cable (DC-jack type.). ====snip==== > Cheers, > Daniel. Officially, USB power is limited to a max, negotiated, of 500mA per socket, with an unnegotiated 100mA limit. In practice, IME (and as far as all the many PC motherboards I've dealt with), the power is simply fed via a 3A polyfuse (or not, in the case of most, if not all, PC Chips MoBos) straight to the 5v pin of whatever number of ports there happen to be available in the I/O shield region, including the 5v pins of the PS/2 sockets (and possibly the MoBo usb headers as well). Notebooks are the most likely place you'd expect USB port power limitations to actually be applied but even here, a simple polyfuse protected 5v bus may be all that is employed to distribute power to the USB ports. IOW, no current limit other than the 3.5 to 4 amp trip of the polyfuse (or whatever the cct trace or smd inductor happens to burn out at on a PC Chips MoBo ;-). Assuming you're plugging into USB sockets that strictly enforce the current limits, the max your "2 x USB" plugged lead can supply is 600mA (500mA negotiated on the real data lead and an additional 100mA unnegotiated from the 'extra power' plug). What I've found is that it's the voltage fed from the USB ports that seems to cause the most problem with USB powered devices. Even when it is within the +/- 5% PSU tolerance range (let alone the +/-10% component tolerance range) external USB 2.5 inch hard disks can throw errors, notably when the voltage is on the low side of the range (just 1% low seems enough to cause trouble). Slightly low 5v is more likely in MoBos that use this rail to power the CPU's VRM (Typically, the older socket A boards and their VIA chipsets - probably the main reason for VIA's bad rep with USB powered devices such as those infamous Speedtouch modems[1]). The P4 intel chipped boards had less trouble since they universally made use of the 12v rail to power the CPU's VRM (with some better MoBo designs allowing the on-board 5v rail to be automatically used in the absence of a 4 pin 12v connection). This meant that the 5v rail had a better chance of being in the -0/+5% tolerance range and, therefore, less likely to fall foul of the out of spec voltage requirements of a lot of USB powered devices. [1] It has to be said that what is "infamous" about the Speedtouch modem is the fact that such an obscenity (of usb interfaced madness) should exist at all. As far as the limitations of USB goes, the Speedtouch was actually quite good if its use with Linux drivers is any guide. The problem with the current versions of USB is that it is a kludge of an interface with no pedigree whatsoever and I could quite cheerfully put a gun to the head of whoever was responsible and squeeze the trigger. -- Regards, John. Please remove the "ohggcyht" before replying. The address has been munged to reject Spam-bots.
From: Daniel James on 14 Dec 2008 09:53 In article news:<31303030373730364943097208(a)plugzetnet.co.uk>, Johnny B Good wrote: > Assuming you're plugging into USB sockets that strictly enforce the > current limits, the max your "2 x USB" plugged lead can supply is 600mA > (500mA negotiated on the real data lead and an additional 100mA > unnegotiated from the 'extra power' plug). Good point ... I'd completely overlooked the fact that the 'dumb' power lead would only get (or, I should say, could only rely on) the unnegotiated 100mA. (Duh!) Thanks for your other thoughts, too. Cheers, Daniel.
From: Martin Gregorie on 14 Dec 2008 13:22 On Sun, 14 Dec 2008 14:53:46 +0000, Daniel James wrote: > Good point ... I'd completely overlooked the fact that the 'dumb' power > lead would only get (or, I should say, could only rely on) the > unnegotiated 100mA. (Duh!) > Hmmm, its surprising that nobody has yet made a 'dumb' USB power cord adapter to fool the socket into supplying 500 mA. It would be at least as useful as certain other USB devices, e.g.humping- dog memory sticks, USB-chargeable NiMH AA cells or USB controlled missile launchers. -- martin@ | Martin Gregorie gregorie. | Essex, UK org |
From: Bruce Stephens on 14 Dec 2008 14:15 Martin Gregorie <martin(a)see.sig.for.address.invalid> writes: [...] > Hmmm, its surprising that nobody has yet made a 'dumb' USB power cord > adapter to fool the socket into supplying 500 mA. I have several. I've got a USB to Nintendo DS Lite, USB to PSP, and some others. Those provide purely power, not data. Such things make lots of sense (IMHO) while manufacturers continue making gratuitously different power sockets. [...]
From: Johnny B Good on 14 Dec 2008 13:22
The message <VA.000015f0.8c2c66dc(a)nospam.aaisp.org> from Daniel James <wastebasket(a)nospam.aaisp.org> contains these words: > In article news:<31303030373730364943097208(a)plugzetnet.co.uk>, Johnny B > Good wrote: > > Assuming you're plugging into USB sockets that strictly enforce the > > current limits, the max your "2 x USB" plugged lead can supply is 600mA > > (500mA negotiated on the real data lead and an additional 100mA > > unnegotiated from the 'extra power' plug). > Good point ... I'd completely overlooked the fact that the 'dumb' power > lead would only get (or, I should say, could only rely on) the > unnegotiated 100mA. (Duh!) > Thanks for your other thoughts, too. It was my pleasure to impart my experiences of usb (in particular, laptop 2.5 inch drives in 3rd party USB2 enclosures). The funny thing is, I've had less problems using drives that state 700mA and even 1A requirements than with those which specify 600mA. Make of that what you will. Personally, I think the newer, larger capacity drives are more voltage critical than the older 3 to 40GB capacity 1 amp guzzlers. -- Regards, John. Please remove the "ohggcyht" before replying. The address has been munged to reject Spam-bots. |