enjoymentboy

I know this has been WELL covered before but I can't find the answer(s) I need in the past posts and I'm hoping that some of you all may be able to help.

I've tried this conversion twice now with now luck. The first supply was ify to begin with so I'm not really surprised it didn't work but this was a brand new one (free) and I tested it first so I know it works for sure.

I put a 10 watt 10 ohm resistor from the +5V to ground, 3 ground leads to a terminal post, 3 +12V leads to a terminal post, connected the brown and orange leads (just as they were on the main power connector) and an led and 330 ohm resistor between the +12V and ground. All other leads were clipped back and sealed with heat shrink and all connectons are soldered and heat shrunk. I have no need for any of the other power leads so they have all been clipped back.

When I plug it in I get nothing. No fan, no light, not even smoke or sparks. Nothing.

Any ideas?

Thanks.

Enjoymentboy

vicman

My Feedback: (10)

here are a few thousand words.

If you need someone to talk to pm me and I will give you my number.

Andrew

Did you connect the PS_ON (green) to DC ground either thru a switch or by shorting directly? Unless PS_ON is pulled low, the PS will not latch. I'm also appending a letter I usually send out when failure to start occurs. You seem to have covered the preload and attachment of the remote sensing wire. Make sure neither of your binding posts are touching the case -- this can cause the overload circuit to capture and keep the PS from coming on.

*****************************

Fan bumps are caused by a Latch_ON followed almost immediately by a Latch_OFF. During the Power On Self Test (POST), several things take place, normally within 100 to 500 ms of startup. The PS will check input voltage to ensure that it will be able to maintain output within prescribed levels (supply current is also continuously monitored to detect brownouts), output levels are tested and pulled high in a predetermined order, some designs test for existence of remotely sensed voltage and some designs will check for a preload. If these are within design limits, PS_OK is set to active high (+2.4v to 6.0v) which releases the MoBo processor from RESET mode and allows the bootup process to begin.

The three most common reasons for Latch_OFF in converted PS's, in order of frequency, are:

(1) Failure to detect a preload.

(2) Perceived short detected by the overload circuitry -- this can be also be triggered by certain chargers with large ripple suppression capacitors or chargers being operated under a high charge rate. Design guide specifications do state that the PS should be capable of sustaining power up and Latch with a 5000 uF load present on the 12v rail. An output short circuit is defined as any output impedance of less than 0.1 ohm detected on any output rail.

(3) Failure of remote sensing to evaluate output voltage levels.

If your 5v rail is intact and the preload resistors are correctly wired and not burned out, then (1) can be eliminated.

Be sure none of your output rails are grounded -- in most PS's I have tested, AC and DC are completely isolated with the case tied to the AC earth or neutral. DC ground is most usually isolated from case ground. In many situations, either there is an internal short of one of the output rails or one of the positive voltage binding posts is touching the case. This causes the overload circuitry to capture and shut the PS down.

The most commonly occurring remote sensing is on the +3.3v buss, usually a lighter gauge brown or orange wire. I have seen on a few PS's sense wires for +5v and ground, but this is not the norm. Check your MoBo connector to see if any of the pins have more than one wire attached -- if so, then check to see if these wires have been reattached in your conversion.

andrew

enjoymentboy

This is EXCELLENT. Thanks. I'll dig into it again tonight and use your advise.

About the green wire....hmmm...I didn't "specifically" do anything with it as it is already connected to the ground through the switch on the back of the case. I left it as it was thinking that it would be a nice and simple way to turn the whole until on\off. Was I wrong there?

I'll try jumpering it tonight and see if it makes any difference. As well I did have the presense of mind to actually mark the brown and orange wires that were joined at the 2X12 connector so I know for certain that the correct ones are connected there.

I'll "try" and post a pic tonight as well but the camera isn't too good and it's a bit of a mess int here. I didn't plan on removing any of the unused wires until AFTER I know the whole things works.

Thanks again.

Enjoymentboy

Andrew

What you probably are seeing is the AC earth ground -- green coming thru the power cable and screwed to the case itself.

There should be a green wire (PS_ON) running to the MoBo connector (pin 14 on the 2x10 and pin 8 on the newer 2x12). This line is held high while the PS is plugged in but not running. Pulling this line low (grounding) is the action that initiates startup. You want to keep your AC and DC separated. [link=http://web2.murraystate.edu/andy.batts/ps/TheFIX_java_table.htm]This link [/link] shows the location of the PS_ON line on a 2x10 (20 pin) MoBo connector -- mouse over the image to see the pin numbers and colors. Either ground this wire or connect it to DC ground thru a toggle switch -- the switch on the back (assuming you have an ATX and not an AT PS -- if it had a remote sense wire attached to one of the 3.3v pins, it will be an ATX) is the master AC switch.

enjoymentboy

AH HA!!!!!! That makes PERFECT sense. I didn't even LOOK at where that green line was going. I just saw green and left it at that. I didn't even really look at any of the other wires that I didn't use (because I wasn't using them[:@]).

Thinking about this now I am pretty sure it'll work fine when I get home.

Thanks so very much for the insight. I'll let you know how it goes.

Enjoymentboy