Hello, I am currently thinking about adding some circuitry to my Linksys Wireless-G USB adapter and was wondering if it was possible to supplement/power the unit independently from the bus. My theory so far is to cut the postitive voltage wire of the USB connector on the unit and supply, via a wallwart a 5.0v power source. Thus the only wires that feed back to the computer, from the unit, are the data lines and ground (because it is needed for reference). Would an idea such as that work? Thank you.
Use the 5v from your PSU its rated at something like 20A+ anyway. USB by design hs a 500mA limit ( i think). Plus 'wall warts' dont usually give that much current.
That's not the question being asked. Question is, will the controller recognise the device when its plugged in. I can't really see why it won't work. Some scanners, etc, use the USB supply, others their own power block. If the USB controller needs to sense some current, add an indicator led. Not much help, but some interesting reading here.
I don't like the idea of the 5V line being left hanging. I'd put a small resistor on it. On most (all?) USB chips supply current for each connector is monitored, so it might get confused if there was no current whatsoever on that line. Also, I wonder if a load on the power line is sometimes used to detect the presence of a USB device so that it can be initialized.
Thank you all for the input. I am reading the article that Cpemma posted and it is quite interesting. nleahcim: I never considered adding a load but that seems a very logical thing to do. I should be attempting this within the next couple of weeks and will try connecting my Linksys to a hub first to provide a buffer in case anything goes wrong. The circuitry I am thinking about adding is a fan that will be powered by a DC-DC converter (also to be built), a VU-type meter (basically the activity LED fed through a freq-volt con. and ADC), USB to ethernet cable converter (I have to transport the signal about 40 feet), and the obligatory extra LEDs. Thank you.
http://www.siliconchip.com.au/cms/A_102685/article.html Copy the link into Google and click through from there to get access to the full article. The design is probably overkill, but should be helpful.
Ok here is my take on this. Run the +5v from a molex to the +5v line on the usb Run the ground from the molex to the ground of the usb Done. The +5v shouldent feed back into the psu or usb port but you could put a diod before where you have connected the +5v from the molex just in case. Same for the ground side. This is how I'm planing on powering all my usb devices in my carpc unless some one gives a sound reason that it wont work.
A diode, even a schottky, may lose too much voltage if your 5V rail is slightly low to start with. A low RDS MOSFET switch loses much less at low currents (under 1A), you could probably use the PC's 5V to supply the "hub" and eliminate the 7805. Another option would be to use an external supply of 6V or more with one of the 5-pin 5V regs that has an 'enable' pin, eg Micrel MIC29151-5.
Is there even a need for the diods? From what I can tell you don't. Ok lets say that wont work then you could tap power from the 12v and use a regulator and step it down to 5v. I don't know if they make any 2 or 3 amp 5v regulators that work using duty cycle insted of just wasting the extra voltage as heat.
Many circuits use a few extra parts just to safeguard against "what-if" conditions. Looking at your basic idea, Code: 5V----------+--------+ | | | 5R USB5--->|---+ | | 0V-------------------+ Suppose the auxilliary 5V supply is cut off for any reason? USB port is suddenly faced with a (5-0.6)/5 = 880mA load. Oops! I think both Maxim and the 'Silicon Chip' circuit's author understand what's needed to meet the USB spec safely far better than me, I'll go with their methods.
I'm 99% sure that the usb controler would realise that it can not supply that much to the device and would thus just disconect it. I gave it a try with my TV Walker Ultra (takes some thing like 1.1 amp at 5v). The special cable has two usb A plugs on one end. One with 4 wires for power and voltage and the other just has the outer two pins for voltage. I just pluged both in and then removed the plug with two wire and sure enough it just disconected the device. I will see if I can find any thing in the usb specs about it though. Also if your using the 5v rail from you psu then there would never be any chance of it cuting out and the usb still runing because if you lose your 5v rail your pc will shut off.
I had a few more thoughts on this subject while I was ordering parts to make my "modded" USB-wireless unit. The main reason I am adding circuitry, in the first place, is to be able to extend the wireless transciever about 40ft from my laptop since my college's wireless is not strong enough to reach my bedroom. My concern is whenever my laptop is powered down, the unit will be always on. What would happen if the unit tried sending data to the laptop? Does the USB circuitry, whenever the host is turned off, have a chip enable on-board to avoid feeding a logic signal to the reciever buffers without power? I thought about circumventing the problem by having my own buffer chip in between the USB unit and the laptop to break the circuit whenever the host is off, but this is USB 2.0 and I do not think I can purchase buffer chips that will be able to handle that kind of switching for cheap and in very low quanity. I could limit the data signal's pathway to the junction where the USB is boosted to be able to travel down ethernet cable, but that will place the active repeater's circuitry in the position of having logic signals applied to it without main power. Another way is to add a power transistor switch between the power bus that feeds into the USB-unit and the wallwart to be able to switch on only when there is power being applied to the USB bus. Therefore, data will only be transmitted from the USB device whenever the laptop's boot sequence allows power to be drawn from its USB ports. I obtained this idea by watching my USB hub, even if it is plugged in, power down whenever my laptop powers down and come back up only during the OS loading of the laptop. What are your thoughts? Thank you.
As far as I know there should be no issue of your usb device waking your pc up. There may be special commands over usb that will wake up the pc but I do not think I have ever noticed this with any usb wifi cards that I have used.
