problem of LOW voltage on USB ports of Intel 915GLVG mobos

[mvs_sarma]

Of late I had a need to measure the voltage of Vcc on USB port of Intel 915 GLVG mother board.
presently 2 ports are used for printer and for optical mouse.

when a third device is connected for programing purposes, i found to my dismay, that the voltage falls to 4.44V or at times even 3.85V.
we generally expect 5V
I need help on how to sort this issue.

[Darrel Taylor]

An easy fix might be to use a "Self-Powered" USB HUB.

Usually they have a DC "Wall Wart" to provide power for more devices.

[mvs_sarma]

An easy fix might be to use a "Self-Powered" USB HUB.

Usually they have a DC "Wall Wart" to provide power for more devices.

I get your point, but the facility of USB auto detection will not ne anymore effective, I am told.

Instead i can take power from one of the spare connectors of the SMPS of the PC itself.as it is measuring 5.14V approx.

[mister_e]

i'm curious to know how much current all USB device connected needs and how much your actual PSU can provide. Seems like a current protection, unsufficient or 'about to fail' psu to me.

Check if your mother board have a Jumper for USB power. some allow you to change from Vusb to 5V (more specific term... but i never remind which one)

EDIT: I've look to you Mobo manual.. yours don't seems to have that USB power source jumper

[mvs_sarma]

i'm curious to know how much current all USB device connected needs and how much your actual PSU can provide. Seems like a current protection, unsufficient or 'about to fail' psu to me.

Check if your mother board have a Jumper for USB power. some allow you to change from Vusb to 5V (more specific term... but i never remind which one)

EDIT: I've look to you Mobo manual.. yours don't seems to have that USB power source jumper

Mine is arond 2 years old PC. I made a check of the situation. the actual %V DC is right at 5.12V at many load points. At the same time the USB derived voltage is around 4.44V or even 3.84V at times. I initially suspected SMPS. But it is OK.

I fear the supply to USB is given in series to any sensor for the current and thus the sensing element drops across it. I wonder, whether the internal structure of Intel USB driver chip is to be studied.

[mister_e]

yeah sure the on-board USB have a current sensor, not sure if a PCI card have it... if so, how better worst it is? I'll measure it here.

As Darrel says, an external powered USB hub 'may' solve the problem. i don't get your point of auto-detection...

But for sure, there's a current limiting somewhere, XP should warn you in case you overload it.

EDIT: Measurement done here...

My HP machine
Load 5 Ohm..
On board USB - 4.26 volts,
PCI Cards 4.1 volt

My intel machine
On-Board 4 volt

[mvs_sarma]

yeah sure the on-board USB have a current sensor, not sure if a PCI card have it... if so, how better worst it is? I'll measure it here.

As Darrel says, an external powered USB hub 'may' solve the problem. i don't get your point of auto-detection...

But for sure, there's a current limiting somewhere, XP should warn you in case you overload it.

EDIT: Measurement done here...

My HP machine
Load 5 Ohm..
On board USB - 4.26 volts,
PCI Cards 4.1 volt

My intel machine
On-Board 4 volt

Sir, please see the pdf form Intel titled "Power Delivery Design Issues for Hi-Speed USB on Motherboards" i got it from www.usb.org please try to put the article title on google , you land straight on the location .