Ramps disconnects when power is applied

Hi I have a ramps 1.4 board with an SD card add on. I've been running it for about a month now.
This morning there was a power cut in my neighbourhood and since then the board has been behaving strangely.
I can connect using repetier host and monitor the temperature of the hotend and bed accurately (checked against an infra-red thermometer). I can also access and change files in the sd card.
However when I turn on the power, the device immediately disconnects from my PC.
I've checked the resistance on the heating elements fans steppers etc. to see if perhaps something shorted during the outage but they are all fine.
I have a spare board kicking about which I wired up, but it exhibits the exact same behaviour, even though it was sitting on my shelf when the power went out.
My power supply is a hacked PC supply, it is delvering about 11.95v
Any idea what could be wrong?

I was poking around with the board a bit just now, I managed to get the motors moving with the power applied to the board, but when I tried to heat the hotend, it disconnected again. could have something to do with the current required for the heater vs the motors? I haven't been able to replicate this, seems like a one off.
I'm wondering if maybe something on the arduino could have blown? I have a spare arduino as well but didn't think that could be part of the problem. (edit: I tried te spare arduino just incase but the same thing happened, connected fine, accurate temperatures sd card access, connected the power and it's gone)
I've ordered some diodes so I can add D1 which is not installed on either of my boards, maybe that will help?

Edited 2 time(s). Last edit at 06/20/2014 11:57AM by blabberjack.

Have you completely disconnected your PSU from power and left it for like 10 minutes to discharge? I've seen some cheap PSU's do weird things after power issues, and only letting them completely discharge seems to help.

I tried that just now, I managed to get the motors moving again, but when I tried to heat the hotend, it disconnected the board. cruisouly the LED indicating that the mosfet for the heater was saturdated stayed on and the hotend heated up, the board was just unresponsive.
I don't have an alternate power supply but if my experiement with the diodes fails that is what I will try next.

I got the diodes today and soldered them in.
I can now get the motors moving much more reliably, I power up the board, reset it and then plug in the USB cable. This gets me connected to the board about %50 of the time. While the board is powered in this manner, I can move the steppers around and turn the fans on and off. However I am still finding that my board loses it connection to my PC when I heat either the printbed or the hotend. I might try an SD card print since then it should continue even if the computer loses connection but I am not hopeful. I would really like to get this solved, hate to think that I fried two boards for apparently no reason.
Just to clarify the PSU is not affected when I attempt to heat the hotend or printbed, it continues to run and heats up the part so it is definetly not a short which would turn off the PSU (I've shorted that supply often enough mucking around with it so I know this behaviour very well).

Ok, turns out that printing from the SD card does work, even though the board will immediately lose connection to the PC when it powers the heaters.
If anyone else has this happen I would recommend disconnecting the USB cable from the board as soon as the then "run print" command is sent. I didn't do this the first time and my USB hub, through which my Ramps connects to my printer, went dead. After I unplugged it and plugged it back in, it works fine but I'd hate to think what might have happened if my board was plugged directly into my PC.
Maybe having an LCD would remove these issues. I may get one if this rather unusual workflow becomes a big nuisance, it is quite annoying not being able to tell how far through the print you are or check temperatures etc. etc.
It's a shame someone with a bit more knowledge of electronics hasn't chimed in, I can only assume some component has been damaged on the board such that when a large current is applied, some makes its way back through the usb connection to my pc which terminates the connection in response, I'm sure if you knew what to look at it would be easy to fix but my crude solution seems to be working for now.

Urgh. Hopefully this will be my final update for this thread.
I managed to get a couple of complete prints out with the fix previous discussed. One completed, one aborted about 95% done and then after that I found that the board would again sporadically lose power, therefore resetting. Sometimes this would happen before the bed got up to temp. sometimes it would happen a few layers in etc.
I concluded that either everything was broken beyond repair or that it was my power supply, I've since replaced the hacked ATX with an OEM PSU I picked up from ebay for about £11 and for the moment it seems to have solved my problems, I can now keep my PC connected to the board during prints and everything seem stable.
Probably could have avoided all of this hassle if I had invested in a decent multimeter years ago, d'oh well.

Still going after abut 20 hours of printing, seems stable so I guess it was the PSU all along.
I wish I could recommend people experiencing this problem troubleshoot the output from their PSU first, but honestly I don't know how you would do that. One thing to bear in mind though is that of all the electronic parts, the PSU is the cheapest to replace so maybe if you encounter an issue like this, it would be worthwhile swapping out the PSU if there is no obvious damage to the board (might also be a good idea for someone with the know-how to put a section in the ramps wiki suggesting components/ traces to test and what to look for if you suspect a fault with your board). Sorry I can't just go and do that myself, I just don't have that experience!

Quote
blabberjack
Still going after abut 20 hours of printing, seems stable so I guess it was the PSU all along.
I wish I could recommend people experiencing this problem troubleshoot the output from their PSU first, but honestly I don't know how you would do that. One thing to bear in mind though is that of all the electronic parts, the PSU is the cheapest to replace so maybe if you encounter an issue like this, it would be worthwhile swapping out the PSU if there is no obvious damage to the board (might also be a good idea for someone with the know-how to put a section in the ramps wiki suggesting components/ traces to test and what to look for if you suspect a fault with your board). Sorry I can't just go and do that myself, I just don't have that experience!

I would have sugested to not get another ATX PSU. They are really not designed for this kind of thing in mind. They will overheat or droop. It seems like most of us on this board, we are all price concious, May I suggest a second hand server PSU or if your willing to invest, a quality dedicated 12v psu? I cannot stress this enough, quality power is quite important. I used to build gaming machines, and have seen "Quality ATX" psus degrade over time. Anyone complaning about power instability seems to either use ATX or use cheap no brand psu's that droop when you turn on a microwave.

---

My Personal Blog. Build blog.
[engineerd3d.ddns.net]

Modicum V1 sold on e-bay user jaguarking11