need help tracking down where this short could be Gen7

Ok the following even though 5V is not attached (power supply isn't either) when I connect the pcb to my usb serial convertor it lights up the standby led
and if I connect an atx power supply even though the atx jumper is the one jumpered the power led lights up and that shouldn't do this unless the psu turns on.
made some changes to the board but that was only to save a few resistors and jumpers and it shouldn't cause this.
please take a look and give me hints where I should be looking for the short.
Regretably I have no digital camera right now or I would blow up a photo and look for it that way
Help
terramir

If the USB converter is powered then its receive output will idle at logic 1. This will feed current into the RXD pin of the ATMEGA and try to drive it to ~5V. That will forward bias the static protection diodes and try drag VDD up to ~4.3V. Depending on how strong the driver is and how much load there is on VDD it can be enough to light LEDs.

Probably not a board fault but a design fault connecting an un-powered device to a powered one without taking this into account.

---

[www.hydraraptor.blogspot.com]

Thanks nophead for the info,
But what about the other led I read both my and the other schematic and from what I see that's connected to the 5V not the 5V vsb that should not light up unless you give an M80 command, the whole reason for asking this is because the whole board is fubar. I could parallel program the fuses and the boot loader but had to put it in a different board to program marlin wouldn't flash on this board so I'm thinking short or I f'ed up on the redesign. but comparing these two I dun see how I could have fucked up functionally they are identical unless I'm missing something.
Please take a look folks
terramir

one thing I found out just now it's actually the firmware turning the psu on right off the bat, maybe this is causing the flash programming to go haywire
About three to four seconds after the red light comes on standby the blue (power) comes on. Maybe I got a pin.h issue or something I gotta go over the configuration becuase the older version of marlin with a 1.3.1 board didn't do this
terramir

turns out that the blue power led is switching on because the psu turns on, because of something :s however unless I fraked up my modifications I can't figure out what the problem is because I looked at the pins.h and well it's on pin 15 and the power supply turns on after 3 seconds and it only does it when the chip is inserted I took high res photos of the board and as of right now I have not found a short that would cause the at mega to act erratically. so unless the library for the gen7 is incomplete in some way for the 1284 or the marlin main v1 has a bug I dunno what is causing this, however I do know that the thing times out on upload and won't connect to pronterface.
It could be that two things are happening at once #1 the psu is turning on #2 the chip while in linux seems to support 250k baud only actually supports 128k (like in windows and it's still programmed for 250k so I can't connect because I fucked up the flashing best thing I can do right now is reflash the bootloader with the gen7t board I have and then load the firmware again this time with a 115000 baud rate and then retry on my new board. of try to connect to it with the ftdi chip I do have which I know supports 250k I'll have to go through the firmware to see if there is some other kind of setting I fraked up
attached is a high res pic of the circuitboard maybe someone can point something out I missed.



terramir

What firmware do you currently have written to the chip? That sorta sounds like the behavior of the gen7 test firmware, except the heater light should be flashing. Have you tried monitoring the serial port from the arduino ide while running the test firmware? What OS are you using, what programmer, etc?

Also, I'm no expert, but your board looks to me like you need to be more liberal with the flux and ease off the solder. I don't see anything obviously shorted, but it looks like you have quite a few cold solder joints, especially around the heater mosfets.

Edited 1 time(s). Last edit at 05/30/2013 06:42PM by Rezer.

well I got a 664p to work as far as that goes. I flashed it with the 00.23 arduino but with the 1284 all I see is no dice at this moment. also strange I got it flashed with my cp2032 chip but I somehow could not get it to work with the ftdi chip I use for my prusa I dunno what the problem with that is I'll have to check it out further as I move along. Also I'll have to order some 1284p chips and see if I can make that work. but the biggest problem I see right now is that I can't get any chip decently flashed with any of the 1x arduino things so that might be a barrier in using a 1284p chip.
I'll have to figure that out as time moves along
terramir