5V at the USB port without connection to PC

Hi Did you watch the following? http://www.arduino.cc/cgi-bin/yabb2/YaBB.pl?num=1286627335 Please write your thoughts. Kyoji

I'm seeing a similar issue in the opposite direction. When connected to USB I am seeing voltage on Vin.

Yep, I see exactly the same thing. I have a high-current draw external load (HL1606 LED strip) that I'm powering with an external 5V PSU, which also supplies 5V to Vin. If I forget and disconnect the PSU whilst leaving the USB cable connected, the Vin pin supplies 5V back out to the LED strip, but it can't supply the current needed (up to 2A), and the strip goes haywire. I haven't toasted my Duemilanove yet, but trying to draw 2A from it clearly isn't a good idea. Seems like a design flaw to me.

Hi Alan

I wrote the same thing in two places Chigau words. Because both write, and write your response here and there.

555 has an internal network with 100kohm The window comparator. (18kohm Another theory says).

Power and network they are connected to GND, 2 / 3 of the "control" is terminal. Meanwhile, the "trigger", "threshold" between networks and terminals 100kohm Equivalent circuit will have no hysteresis.

Used to detect when the control terminal, the protective resistance 100kohm used. The temporarily stable, Network "3 / 3-2 / 3" Resistance and 100kohm 50kohm and parallel equivalent.

Therefore, the "threshold" terminal "3 / 3-2 / 3" is close to the resistance 51kohm used.

Instability because the show circuit simulator, 1nF capacitor connected to both. Please try to prove too.


I found the same behaviour of the Vin pin like James and Alan.

At a similar application like this: http://www.david-laserscanner.com/forum/viewtopic.php?p=12366#p12366

....a permanent connection via USB is necessary to send motor trigger impulses in synchronism with the captured camera frames. I noticed that even if the wall power supply wasn't plugged in the laser and the stepper motor could be powered from Vin. To avoid this lack I have inserted a diode (1N4001) in front of the Vin pin. (Of course kathode to Vin). A test of Vin as an output revealed: Without any load >> 4.5V At I = 42mA >> 4.2V I = 100mA >> 4.1V I = 180mA >> 3.9V I = 360mA >> 3.6V

But this effect has nothing to do with the wrong replacement of the FDN340P for the NDT2955. It is caused by IC4, the voltage stabilizer MC33269D-5.0.

In the meantime I have removed the FDN340P from all my boards and substituted them by the old and fine working NDT2955.

Some posts above Gianluca Martino has promised to send results of his tests. Unfortunately he has forgotten us users. :(

n the meantime I have removed the FDN340P from all my boards and substituted them by the old and fine working NDT2955

I wish I could find a NDT2955 to do the same. :(

James, I ordered my NDT2955 from RS Components: http://de.rs-online.com/web/search/searchBrowseAction.html?method=searchProducts&searchTerm=ndt2955&x=32&y=14

And I think they have affiliated companies or distributors all over the world: http://www.rs-components.com/index.html

Thanks Walter. I don't know why I didn't notice before, but Digi-Key did have them in stock. I thought I had already checked there and didn't find them. Order placed, looking forward to making the change.

Hi dude, I've same problem with 5V at gold fuse without USB cable supply. But one of the UNO board is doing well, then I inspected the board and found different usage of LMV358 and LM358. So the UNO with LMV358 is doing well. This IC is supported to be rail-to-rail output but LM358 does not. I'm still looking for the replacement of LM358.

Hope it will help.


PS: New UNO is not using 16MHz crystal anymore. both are resonators.

Would a USB cable with the power wire cut work around the problem? You could then power the Arduino Uno from an external source while still talking to your PC, without modifying the board itself.