Help! ATX PSU for i3

Hello fellow RepRappers!

I've finished my i3 build and I'm very excited to start configuring my rig and make my first print. However, I'm having problems with my power supply.

I had a few brand new 500W ATX PSUs laying around for another project I'm working on, so I figured I would covert one using these directions: [reprap.org]

The only issue is, it doesn't work! No voltage on the multimeter, no 12V light on. I'm assuming I've wired it incorrectly, so I've attached a few pictures to this post. If any of you have experience with this, would you please help me trouble shoot?

Pics:
1 - The Green "PSU ON" wire is connected to a Black GND wire.

2 - The Brown 3.3V Sensor wire is connected to the Orange 3.3V wires.

3 - The Red 5V and Black GND to the 10ohm 10W resistor.

4 - Wiring to the binding posts.

5 - Grey "Power On" wire and the Purple 5V standby -- these are not connected!

6 - Remainder Yellow 12V wires not wired to binding posts, connected together and insulated.

7 - Remainder Black GND wires not wired to binding posts, connected together and insulated.

8 - The PSU switch is on in this image. No fan movement. No light.


Please let me know what I'm doing wrong. Thank you in advance!

I think the problem is the purple and grey wires. The grey is power ok signal. Meaning it is switched high when power is stable. Connecting it to 5vsb(purple) both unnecessary and I unadvisable. Try just disconnecting it.

Edited 1 time(s). Last edit at 01/24/2014 02:25AM by lazzymonk.

Hi Lazzymonk,

Thanks for the reply, those wires are actually not connected together, they're both insulated, I just put them in an endcap together to save space.

Does that mean the rest of the wiring looks good?

Could it be that this PSU has some circuitry to circumvent wiring like this?

Thanks again!

only other thing I can see is it looks like your binding posts are connected to the metal case and not insulated from it? That would stop it starting up if it is the case.

Checked this on the multimeter, seems to be okay (the case is powder-coated and isn't conducting).

After turning it PSU off and unplugging it, I pulled an OL reading from the main terminals, which eventually dropped to zero. Seems like the device is definitely drawing power charging the capacitors, it's just not pushing it out to the terminals.

Here's a link to the PSU: [www.logisyscomputer.com]

Manufacturer says it's got "Over Voltage Protection, Over Power Protection and Short Circuit Protection." I'm guessing, based on the fact that it is charging the caps, that I've done some faulty wiring which is tripping one of these systems...

Might have to tear it apart over the weekend and start all over...

I would pull the cap off the 5vsb(purple) then and see if it has 5v on it. If it does not then I would say something is shorting somewhere.

Last time I had this issue, I neded up replacing with a known good motherboard.

Assuming that the PSU was working OK before you started conversion (you did test it right? ), about the only thing can be wrong is a short on the output. ATX psus are pretty good at not starting if there is a short on the output (I have tried it). I would undo the terminal posts to check.

Unless the terminal posts are fully insulated, I would not bolt them to a metal case, even if the case is painted. I'm not terribly keen on the PSU conversion idea anyway, it's a bit unnecessary and error prone. Does depend on people "knowing what they are doing".

I am sure I saw an ATX adapter/breakout board somewhere which does a similar job and is easier to use.

---

What is Open Source?
What is Open Source Hardware?
Open Source in a nutshell: the Four Freedoms
CC BY-NC is not an Open Source license

Atx breakout boards are easy enough to make with very few components. I have made a few in the past.

If you disconnect all the wires then touch the green wire to black does the psu start?
Also a simple mistake that I have made before - the power switch on the back of the psu in on right?

Hi everyone,

Thanks for all your responses and suggestions - I've got it working!

Turns out upon further investgation that, due to the binding posts being screwed into the frame, it was shorting out after all. Good learning experience -- test twice before I open my mouth!

I cut the whole thing up and started over -- same wiring, same binding posts, but this time I enlarged the holes and covered the screw part of the binding post touching the frame with a bit of heat shrink. Attached is an image of the 12V lamp working.

Going to get some dinner before I plug the i3 in... not sure what will happen after that, but I'm sure I'll be posting back here for help in no time.

Happy printing!

Polymathiae
Twitter: @polymathiae