USB

Too many (self included) have had or still see USB problems, I guess too often it gets blamed, on mystical `non USB-2 compliant cables' , which might occasionally even be true, if perhaps some cables aren't screened too well, or are too thin, but I think for many things there's a much simpler explanation: Inadequate power & bad connections.

USB spec. allows devices to draw no more than 0.5 Amps = 500 mA.

But devices such as external disks Lie & say "self powered", even when no power supply is connected. What this must mean is "Device is capable of or should be self powered" - but does not have intelligence to realise currently Not self powered. ... & sometimes such disks may work, & sometimes not ... (what's the betting more likely if the hub power is plugged in ;-)

Power consumption varies. If your computer or hub power supply is strained, (or cable kinked & some wires fractured to make a high resistance) not all devices may be affected the same way.

Example: Some USB memory sticks power ratings from `usbdevs -v`

• 256 M - Celldisk 100 mA
• 512 M - Kingston 100 mA
• 512 M - MyFlash 80 mA (stick now dead & gone)
• 1 Gig - Integral 200 mA (this has a 1.4M pretend floppy lock/ file system, maybe using more high power logic ?)
• 2 Gig Sony - (Many were Defunct) 100 mA
• 2 Gig SanDisk 100 mA

Other devices

• Havit USB Keyboard is rated on back as 0.25 Amps
• USB to PS/2 converter "Semi Tech PS/2 Keyboard - PS/2 Mouse" reports with `usbdevs -v` 100 mA, regardless if mouse & keyboard plugged into it.
• Larger Havit PS/2 keyboard (that I use through a USB to PS/2 converter is not labelled for power, so no idea what it draws
• A small blue Kensington mouse isn't power labelled, but `usbdevs -v` reports 100 mA
• A mouse given away as a pharmaceutical promotion reports: "low speed, power 98 mA, config 1, USB Optical Mouse(0x00c3)" (nice to see a non round number :-)
• [ I'll solder up a socket to monitor power consumption later. ]
• ( Internal mouse pad inside laptop reports 100mA. )
• Disk: Toshiba MK1403MAV 12.7mm IDE 1.44GB Paper label say 0.7 A.
• MK6021GAS HDD2183 MQ60kfGG 9.5mm IDE 16 16,383 2 1 60GB 5V 4.7W on startup, 2.6 Watts seeking, 2.3 R/W

I've just confirmed (as expected): That two ultra thin usb cables, on rewinding drums, wont pass enough power to an external disk (small laptop disk in small box, not a big chunky 3.5" disk) to start disk spinning (but will supply enough power to keep disk spinning if already started), So external power needs to be supplied direct to disk from transformer (even though thin cable carries sufficient power to turn on LED that doubtless works at lower voltage, & too high a cable resistance to supply enough power than disk motor & drive & adapter electronics. Confirmed as when wobbling it, I could not feel disk spinning.

With external power direct to disk, then it works (& one can feel it spinning too). Reason of course must be cable is simply too thin for disk power. Same disk works without direct power, if using a bit thicker USB extension cable to a powered hub.

The disk was not recognised (no /dev/da* on FreeBSD) unless the hub was powered up, or external power for the disk

This all at USB-1 speeds, so no mystical speculation about magic of USB 1 v. USB 2 cables applies ;-)

Of 3 USB power supplies:

• 2 came with 2 port USB 2 card from Belkin.
• 1 Manufacturer DVE is rated 5V 2.4 Amp. Plug is too short to fit reliably in the Belkin Cardbus (credit card size) USB adaptor. Plug is far too short (7.5 mm) to do much more than wobble quite unreliably in drive box garyj@ gave me. Plug was long enough to fit in a Belkin 4 port USB-1 hub though. I cut off 2 mm of plastic on end of plug.
• Other PSU looks identical shape but labelled Achme Corp, rated 3.0 Amp & Longer plug, 9.5 mm, which sits well in Belkin USB cardbus adapter & in garyj@ drive box. Belkin card has 3 A written on it.
• The transformer of Zediworks 4 port USB2 hub is labelled 2.1 Amps Zediworks (subd reports an NEC chip) Hub also falsely reports "self powered" even when no power is plugged into hub or disc
• All are 5V, positive centre pin, though different size round plugs.

Small plugs: One should also remember with these smaller than normal size USB plugs, that they're not as rugged as the full size ones, if they wobble the power may get interrupted to the disk, the disk may then reset confusing the Op Systems.

One small USB plug I had was too short: 6mm instead of 7 mm It didn't fit properly into external disk chassis (chassis from garyj@, with Grey plastic fascia) Of course thus a potential double problem of failing power & failing USB signal. Cutting off 1 mm of plastic made it work.

Sometimes the USB hub locks up (crashes). Same solution as anything else, removing external power & USB cable to laptop (& drive if powered I guess) seems to reset (reboot) it.

A power related problem. Likely it could do similar to other OS's eg Linux & XP.

On both full size flat (A) plug & square (B) plug, 2 of the pins (outer 2 on A) are longer to stabilise power before devices can signal on other 2 wires.

---

USB port on main boards

Pin	Name	Description	Cable Colour
1	Vcc	+5V DC	Red
2	D-	Data -	White
3	D+	Data +	Green
4	Gnd	Ground	Black

---

WARNING: SITE IN .RU DOMAIN Ref: http:// pinouts .ruWARNING: SITE IN .RU DOMAIN Local rooted file://	WARNING: SITE IN .RU DOMAIN Ref: http:// pinouts .ruWARNING: SITE IN .RU DOMAIN Local rooted file://
Plug at peripherals. (B) Plug on cable, from single upstream side of a hub to a PC.	Socket (jack) at controller. (A) As on a PC chassis & 4 outputs of a Hub. (B) Single Socket on chassis of a 4 port hub

---

Ref usbman.com

Wire Colour Code Other
COLOUR:	CONNECT TO:
Red, Orange	Power, VCC, SVCC, +5V
White, Yellow	P-, Data-, SBD-
Green, Grey	P+, DATA+, SBD+
Black, Blue, Brown	Ground, Grnd, Gnd

---

Cabling